The English Consonant Mispronunciations

THE ENGLISH CONSONANT MISPRONUNCIATIONS PRODUCED BY SUNDANESE NATIVE SPEAKERS AN UNDERGRADUATE THESIS Presented as Partial Fulfillment of the Requirements For the Degree of SarjanaSastra in English Letters

By ALOYSIUS PRIANTO RAHARJO Student Number: 054214017

ENGLISH LETTERS STUDY PROGRAMME ENGLISH LETTERS DEPARTMENT LETTERS FACULTY SANATA DHARMA UNIVERSITY YOGYAKARTA 2010

By ALOYSIUS PRIANTO RAHARJO Student Number: 054214017

ENGLISH LETTERS STUDY PROGRAMME ENGLISH LETTERS DEPARTMENT LETTERS FACULTY SANATA DHARMA UNIVERSITY YOGYAKARTA 2010

iii

FOR

MY BELOVED PARENTS

IN HOPE OF A BETTER FUTURE

vii

ACKNOWLEDGEMENTS

I would express my great thanks to Jesus Christ for His blessing and almightiness, give me the strength and spirit to finish this undergraduate thesis.

I also would express my countless thanks to my parents, brother, and sister for the material and immaterial support. Without you this undergraduate thesis would be never done.

I would like to appreciate my gratitude to Dr. Fr. B. Alip, M.Pd., M. A. who has patiently and wisely guided me from the beginning until the end of the making of this undergraduate thesis.

I would extend my gratitude to Drs. H. Kasiyo M. Pd. as the headmaster of SMAN 1 Kuningan, Jawa Barat for the permission given to do the research at the school. And also Dra.Ii Wasita M. Hum.and Drs. AgiAgusMulyadi M. Hum. as the English teachers of SMAN 1 Kuningan, I would appreciate my gratitude for the help given during the collecting the data in the school. I do not forget to express my gratitude to the nineteen students of SMAN 1 Kuningan whose names can not be mentioned here one by one for the willingness and time to have their speech recorded.

Last, I would like to express my thanks to my friends DoniAgungSetiawan, AgathonHutama, JhoniFrederikusSikantLayang, BayuDewaMurti, and all of my friends whose names can not be mentioned here one by one for the support and inspiration given. Without all of you my life would be not so meaningful and colorful.

viii

TABLE OF CONTENTS

TITLE PAGE...... i APPROVAL PAGE ………………………………………………………... ii ACCEPTANCE PAGE …………………………………………………… iii MOTTO PAGE …..………………………………………………………….. iv DEDICATION PAGE ………………………………………………………. v STATEMENT OF ORIGINALITY ………………………………………… vi PERNYATAAN PERSETUJUAN …………………………………………. vii ACKNOWLEDGEMENTS ………………………………………………… viii TABLE OF CONTENTS …………………………………………………… ix ABSTRACT …………………………………………………………………. x ABSTRAK …………………………………………………………………… xi

CHAPTER I: INTRODUCTION …………………………………………….. 1 A. Background of the Study ………………………………………………... 1 B. Problem Formulation ……………………………………………………. 3 C. Objectives of the Study …………………………………………………. 4 D. Definition of Terms ……………………………………………………... 4

CHAPTER II: THEORETICAL REVIEW ………………………………… 6 A. Review of Related Studies ……………………………………………... 6 B. Review of Related Theories ……………………………………………. 8 C. Theoretical Framework ………………………………………………… 33

CHAPTER III: METHODOLOGY ...... 35 A. Object of the Study …………………………………………………….. 35 B. Method of the Study …………………………………………………… 35

CHAPTER IV: RESEARCH RESULT …………………………………….. 41 A. Data Description ……………………………………………………….. 41 B. Contrastive Analysis of English-Sundanese phonetics ………………… 42 C. Phonological Analysis of The English Consonants Mispronunciation … 44 D. Summary of the Research ……………………………………………… 70

CHAPTER V: CONCLUSION ………………………………………………. 74

BIBLIOGRAPHY …………………………………………………………….. 76

APPENDICES…………………………………………………………………. 78 Appendix 1 …………………………………………………………….. 78

ABSTRACT

ALOYSIUS PRIANTO RAHARJO. The English Consonant Mispronunciation Produced by Sundanese Native Speakers. Yogyakarta: Department of English Letters, Faculty of Letters, Sanata Dharma University, 2010. Mispronunciation is the performance errors or mistakes of pronunciation caused by fatigue and inattention, and also lack of knowledge of the rules of the language. Mispronunciation is also the result of the stiffness of mature speech organs (fatigue), the failure to imitate the sound (inattention), and poor abilities how to produce the sound in the learning process (lack of knowledge). The English consonant mispronunciation analyzed is Sundanese native speakers. There are two reasons to take this study. First, the writer wants to know the consonant differences between English and Sundanese. Second, the writer wants to know the the English consonant mispronunciation produced by Sundanese native speakers as the result of consonant differences. There is a wide gap of consonant inventories between English and Sundanese. There are two problems in this study. The first is what phonesare similar and different in English and Sundanese. The second problem is what are the English consonant phones mispronounced by Sundanese native speaker as the result of consonant differences and how can they be explained phonologically. From these problems, the objectives of this study are to find out the similarities and differences between English and Sundanese, to identify the English consonants mispronounced by native speakers of Sundanese, and to understand the phonological process of the mispronunciation. There are two methods applied in this research to answer the problem. The first method is contrastive analysis which looks for the similarities and differences between two languages. It is applied to answer the first problem. The second method is the error analysis which analyzes the English consonant mispronunciation by native speakers of Sundanese as the result of the language differences. It is applied to answer the second problem. This study found that there were nine English consonants which were predicted would be mispronounced after contrasting the consonant both English and Sundanese such as [f v θ ð ʧ ʤ ʃ ʒ z]. The fact in the field showed that the English consonants mispronunciation often occurred in some consonants like [v θ ð ʒ z], shown by their accuracies were under 50%. The mispronunciation was caused by their absence in the Sundanese phonological system. Other consonants such as [f ʧ ʤ ʃ] were not difficult to be pronounced shown by their accuracies were more than 50%.

ABSTRAK

ALOYSIUS PRIANTO RAHARJO. The English Consonant Mispronunciation Produced by Sundanese Native Speakers. Yogyakarta: Jurusan Sastra Inggris, Fakultas Sastra, Universitas Sanata Dharma, 2010. Salah pengucapan adalah tindakan pengucapan yang keliru atau salah, yang disebabkan oleh factor kelelahan, ketidakperhatian, dan minimnya pengetahuan mengenaita tabahasa. Salah pengucapan juga merupakan hasil dari kekakuan organ yang berfungsi untuk berbicara (fatigue), kegagalan meniru suara (inattention), dan kurangnya kemampuan untuk menghasilkan suara dalam proses belajar (lack of knowledge). Salah pengucapan yang dianalisis adalah konsonan Bahasa Inggris yang diucapkan oleh penutur asli Bahasa Sunda. Alasan dilakukannya penelitian ini ada dua, yaitu, yang pertama, penulis ingin mencari tahu perbedaan bunyi konsonan antara Bahasa Inggris dan Bahasa Sunda. Yang kedua, penulis mencaritahu bunyi konsonan bahasaInggris yang salah ucap oleh penutur asli bahasa Sunda sebagai hasil dari perbedaan bunyi konsonan tersebut. Terdapat banyak perbedaan bunyi konsonan antara kedua bahasa tersebut. Pada penelitian ini terdapat dua pokok permasalahan. Pokok permasalahan pertama adalah apa saja persamaan dan perbedaan bunyi konsonan antara Bahasa Inggris dan Bahasa Sunda. Sedangkan pokok permasalahan kedua yaitu apa saja bunyi konsonan bahasa Inggris yang salah ucap berdasarkan perbedaan tersebut dan bagaimana kesalahan pengucapan konsonan bahasa Inggris dapat dijabarkan dari sudut pandang fonologi. Dari pokok-pokok permasalahan tersebut dapat dirumuskan bahwa objektif penelitian ini adalah mencari dan menjabarkan kesalahan pengucapan konsonan bahasa Inggris oleh penutur asli bahasa Sunda. Ada dua metode yang digunakan pada penelitian ini untuk menjawab kedua pokok permasalahan tersebut. Untuk menjawab pokok permasalahan pertama, menggunakan metode analisis kontrastif yang mencari persamaan dan perbedaan antara dua bahasa. Sedangkan untuk menjawab pokok permasalahan kedua menggunakan metode analisis kesalahan yang menganalisa kesalahan pengucapan konsonan bahasa Inggris oleh penutur asli bahasa Sunda sebagai hasil dari perbedaan kedua bahasa tersebut. Penelitian ini telah menemukan bahwa terdapat Sembilan konsonan bahasa Inggris yang diprediksi salah ucap setelah membandingkan konsonan bahasa Inggris dan bahasa Sunda, yaitu [f v θ ð ʧ ʤ ʃ ʒ z].Fakta dilapangan menunjukkan bahwa kesalahan pengucapan seringkali terjadi pada konsonan [v θ ð ʒ z] yang ditunjukkan oleh tingkat akurasi pengucapan yang dibawah 50%. Kesalahan pengucapan tersebut disebabkan oleh tidak adanya konsonan tersebut dalam sistim fonetik bahasa Sunda. Sedangkan konsonan yang lainseperti [f ʧ ʤ ʃ] tidak begitu sulit untuk diucapkan sebagaimana ditunjukkan oleh tingkat akurasi pengucapan yang lebih dari 50%.

CHAPTER I INTRODUCTION

A. Background of the Study

Indonesia has many local languages. One of them is Sundanese, which is spoken in West Java in Indonesia. According to Subarna, there are three major local languages in Indonesia, that is Javanese which is spoken by 75,2 million peoples, Sundanese by 27 million peoples, and Malay by 20 million peoples

(Subarna, 2007; 13). It is interesting to discuss the Sundanese pronunciation problem in speaking English because Sundanese is the second biggest local language spoken in Indonesia after Javanese. According to Syoc, Sundaneseis classified as the Western Malayo-Polynesianlanguage family and has several dialects based on the locations of the people (Syoc, 1959: 1). By means,

Sundanese is the part of Malay language, such as Bahasa Indonesia,use affixes

(suffixes, infixes, prefixes) attached to base words to modify the meaning or to indicate the function of the word in the sentence and pronounce the words less different from its spelling (spelling pronunciation). The areas where is Sundanese spoken are wide-spread in several cities such as Bandung,Ciamis, Kuningan,

Bogor, Cirebon,Majalengka,Indramayu and Tasikmalaya.

As a non-native English speaker who learns English as foreign language,

Sundanese often face difficulties and problems in those aspects. For example, one of my Sundanese friends has pronounced the English word ‘café’ /kᴂfeɪ/ as

[kape]; think /θɪŋk/ as [tɪŋ], sound similar to the word ‘thing’ /θɪŋ/ as [tɪŋ], ‘vase’

/veɪz/ as [pas] and many others. It was funny to hear that mistakes made, but first

we consider that Sundanese doesn’t have affricates and fricative sounds such as

[θ], [f], [v], and [z]. Based on the fact above, it is very interesting to analyze the

English mispronunciation as the result of Indonesian local dialect interference made by the Sundanese speakers when speaking English.

The reason for taking this study is that English words are pronounced differently from its spelling, e.g. the word ‘call’ pronounce [kɔ:l] not *[cal] nor

*[kel]; ‘corrupt’ [kə’rʌpt] not *[kɔrʊp]. On the contrary, Sundanese words are pronounced similar to Bahasa Indonesia, which most words are pronounced as same as its spelling, for example, ‘hudang’ /hudaŋ/ (wake up), tuang /tuɁaŋ/ (to drink), dahar /dahar/ (to eat). Besides, Standard English (IPA) has totally 44 sounds, which consist of 24 consonants and 20 vowels (Indriani, 2001; 8-13);

Sundanese only has 18 consonants and 7 vowels (Syoc, 1956; 50). The description of their sounds will be discussed at chapter 2. So, it is very interesting to analyze the English consonant mispronunciation by Sundanese native speaker because there is a wide gap between English and Sundanese consonant sounds.

The limit of this study is the English consonant sounds. Consonants are easier to be described than vowel by the region of mouth where produced and the way how to produce it as stated by Giegerich:

One of the difficulties of in establishing in the principal features of vowel lies on the in the fact that exemplification with the ‘real-life’ English vowels for two reasons… the vowel…will vary greatly from one accent of the language to another; vowel sounds in English tend not to be as ‘pure’ as we would like the example to be that we use to show the different dimension of articulation (Giegerich, 1991; 13)

To find out the English consonants which are mispronounced by Sundanese speakers, this study tends to contrast both consonant phonetic inventories as the

consonant phonemic representation. According to Peter Roach, phonetics is defined as the scientific study of speech (Roach, 2001:5). So, by understanding the phonetics descriptions of both languages, we can analyze thesimilarities and differences of both English and Sundanese which drives to English consonant mispronunciation when the Sundanese speaking English. In phonetics, we use a particular symbol to represent a particular sound called phonetic transcription which follows the IPA (International Phonetic Association) as the standard phonetic symbol used worldwide. It is a special challenge to find out the English consonant mispronunciation produced by Sundanese through the phonetic gaps and features between them.

Then, this study is very worthy because this study describes both English and Sundanese phonemes clear and specific. It can help Sundanese to adopt an effective strategy in learning English by avoiding the common mistakes in pronouncing English. Besides, this study also discusses the similarities and differences of English and Sundanese phonetics. It can help the Sundanese speaker to understand the problem.

B. Problem Formulation

To support the research of this study, there are several problems to be formulated:

1. What are the phoneswhich are similar and different found between English

and Sundanese?

2. What are the English consonant phones mispronounced by Sundanese native

speaker as the result of consonant differences and how can they be explained

phonologically?

C. Objectives of the Study

From the problems formulated above, there are three objectives in this study. The first objective of this study is to find out the similarities and differences of English and Sundanese phonetics. The second objective is to identify the English consonants mispronounced by native speakers of Sundanese.

The third objective is to understand the phonological process of themispronunciation.

The similarities and differences of English and Sundanese phonetics will be detailed in manners and places of articulation. Thus, the differences of both phonetics are used as the foundation to identify the English consonants mispronounced by Sundanese native speaker. Last, analyzing the mispronunciations which are applied in the words through phonological process.

D. Definition of Terms

There are several terms that are used in this study. They are pronunciation, mispronunciation, native speaker, and interference. To prevent misunderstanding and ambiguity, it will be explained below.

1. Summarizing from Kreidler, Pronunciation isthe process of producing

meaningful speech sound which is articulated by the controlled friction of

airstream through human’s speech organs (Kreidler, 2004: 22-3). So, in

pronouncing any language, the sound is produced consciously by control

our speech organs such as lungs, vocal tracts, tongue, teeth, and lips to get

the precise friction of airstream from the lungs. That is which differ the

human speech from animal sound and infant speech.

2. Mispronunciation is the performance errors or mistakes of pronunciation

caused by fatigue and inattention, and also lack of knowledge of the rules

of the language (Krashen, 1982: 139). According to Krashen, we can say

that mispronunciation is the result of the stiffness of mature speech organs

(fatigue), the failure to imitate the sound (inattention), and poor abilities

how to produce the sound in the learning process (lack of knowledge).

3. Native speaker is one who learned the language before the onset of

puberty. A grammar of language is always based on the linguistic behavior

of native speaker only, since language learned after the onset of puberty is

rarely of the same competence as native competence (Napoli, 1996: 40). It

means that native speaker is the person who speaks certain language from

childhood. It is different when someone learns and speaks certain language

from mature; they do not have equal abilities as native speaker does.

4. Interference is the knowledge of one language (e.g. one’s first language)

negatively impacting on development, production, or comprehension of

another (e.g. one’s second language) (Fasold, 2006: 505). In the other

way, Fasold try to explain that interference is the negative impact or error

sourced from mother language when Sundanese learn and speak English.

CHAPTER II THEORETICAL REVIEW

A. Review on Related studies

This chapter discusses two main sub topics; they are related studies and related theories. In this review of related studies, there are several studies that discussed about non-native pronunciation problem in speaking English. The non- native here is the Sundanese speaker. During the research progress, the problem is the previous study about Sundanese speaker mispronunciation when they speak

English, or phonological contrastive analysis of English and Sundanese, is very rare discussed or even never been done yet.

As the non-native English speakers, they often made a mistake as the linguistic interference of their first language (L1) characteristic and cultural background. Because of many aspects such as the lack of knowledge, different mindset and culture, physical nature, and so on, they transfer their knowledge in grammar and pronunciation of first language and culture context into the target language (L2). Especially Indonesian speakers, they have cultural and linguistic backgrounds that are somehow contrast to native English speakers, as we know the eastern and western culture. As Alip stated in his journal about the feasibility of Indonesian English:

Therefore, the English of the Indonesians is characterized by their physical nature and local culture. As a result, their English might also be influenced by the Indonesian language, both linguistically and culturally(Alip, 2004: 1).

Otherwise, the study of the feasibility of Indonesian English is discuss about how far the linguistic and local culture impact in English pronunciation when the Indonesian speaks English, and the existence of a local English variety so called Indonesian English. Contrast to this study, the focus is English consonant mispronunciation which is produced by Sundanese speaker.

Another study of pronunciation problem also discussed by Thongsin in a journal titled A Study of English Pronunciation Problem for Thai Learners. This study explicitly use contrastive analysis as the basic method to find out the problematic English consonant for Thai learners as stated by Thongsin:

Wherever the consonant /g/, /v/, /θ/, /ð/, /z/, /ʃ /, / ʤ/, / ʒ/, and /r/ occur at aninitial and a final position, the learners will substitute them with the sounds /k/, /w/, /s/, /t/,/d/, /s/, /ch/, /t /, /c/, and /l/, respectively (Thongsin, 2007: 12).

By contrastive analysis, Thongsin found that there are nine English consonant that are absent in Thai. They are /g/, /v/, /θ/, /ð/, /z/, /ʃ /, / ʤ/, / ʒ/, and

/r/. The result of Thongsin’s study is that most Thai learner had no problem with these English consonant. But some of them replaced it with /k/, /w/, /s/, /t/,/d/, /s/,

/ch/, /t /, /c/, and /l/ in certain position in the words.

The statement above underlines the causes of pronunciation problem which is faced by Thai learners when they are speaking English. Thongsin’s research is nearly similar with this study, but it differs in the object of the study that is Sundanese speakers, and the method of problem explanation, that is phonologically not psychologically.

The Japanese also face pronunciation problems when they speak English.

The linguists have identified the problems. One of them discuss about the differences between Japanese and English phonology through contrastive analysis method.

A contrastive analysis of English and Japanese phonology can help to identify potential, even likely, challenges forJapanese speakers of English (JSE). Learners of English will encounter differences in many areas: the distribution ofallophones and phonemes, the number of vowels and consonants, the variety of environments in which fricatives occur,the discrimination and production of /r/ and /l/ and other English approximates, the variety of syllable structures,and prosody (Carruthers, 2006: 17).

Carruthers found that the differences are the allophones and phonemes distribution, the number of vowels and consonants such as English lax vowels and the consonants /ð/,/θ/, and /v/, and the production of /r/ and /l/. This study use similar method with Carruther’s that is Contrastive Analysis, but different in the object of the study.

B. Review on Related Theories

Before we go further in analysis, we need several theories that are appropriate with the topic we are discussing. Here, the topic discuss about the mistake in pronunciation made by Sundanese as the result of Sundanese interference. So, we must understand first the sounds owned by both English and

Sundanese that are very helpful in explanation why the mistakes happen, not just only as a comparison. Besides, the theory of phonetics we also need the theory of interference and phonological rules to provide the analysis.

1. The Description of English Phonetics

In the linguistics study, phonetics is a part of phonology which focuses on the speech sound. As Fromkin states in An Introduction of Linguistics that is to describe speech sound it is necessary to know what an individual sound is, and how each sound differs from all other. It means that it is important to understand how the sounds are produced.

a. English Consonants

According to Fromkin, consonants are produced with some restriction or closure in the vocal tract that impedes the flow of air from the lungs. English consonants have 24 sounds which divided into two types; place of articulation and manner of articulation (Fromkin and Rodman, 2003: 242).

I. Places of Articulation

This term refers to the place where the place in the vocal tract the airflow restriction occurs. In the process of speech producing in our oral cavity, it also needs the movement of our speech organs to the specific place in the oral cavity.

According to McMahon in a book titled An Introduction to English phonology, places of articulation consist of nine classifications (McMahon, 2002: 31-3; see also Giegerich, 1991: 8-12; Jones, 1960; Fromkin and rodman, 2003).

1. Bilabials

Bilabial consonants are articulated by bringing both lips together. So, the articulators are upper and lower lips to produce bilabial consonants.They are

[p],[b], [m], [w], and [ʍ]. Except [w] and [ʍ] are not simply bilabial, but a labial- velar one which is [w] classified as voiced labial-velar approximant, and [ʍ] as voiceless labial-velar fricative. For example:‘put’ /pʊt/, ‘but’ /bʌt/, ‘mutt’ /mʊt/,

‘witch’ /wɪtʃ/, and ‘which’ /ʍɪʧ/.

2. Labiodentals

Labiodentals are articulated by touching the bottom lip to the upper teeth.

So, the articulators to produce labiodentals consonants are the bottom lips as active articulators, and upper teeth as passive articulator. They are [f] and [v]. For example: ‘fan’ /fæn/, ‘van’ /væn/.

3. Interdentals or dental

Interdentals consonants are articulated by inserting the tip of the tongue between the upper and lower teeth. So, the articulators to produce dental consonants are the tip of the tongue as active articulator and both upper and lower teeth as passive articulators. They are [θ] and [ð]. For example: ‘thin’ /θɪn/, ‘then’

/ðen/.

4. Alveolars

Alveolars are articulated by raising the front part of the tongue to the

alveolar ridge. So, the articulators are the front part of tongue as active articulator

and alveolar ridge as passive articulator. They are [t], [d], [n], [s], [z], [l], and

[r].For example: ‘tall’ /tɔ:l/, ‘doll’ /dɔ:l/, ‘nut’ /nʌt/, ‘so’ /səʊ/, ‘zoo’ /zu:/, ‘low’

/ləʊ/, and ‘rural’ /rʊərəl/.

5. Palato-alveolar

According to Giegerich, palato-alveolar is articulated by moving the front part of the tongue to the back of alveolar ridge just a front point of the hard palate.

So, the articulators are the front of the tongue as active articulators and the back of alveolar ridge just a front point of the hard palate as passive articulators. It is clear that palato-alveolar consonants are an alveolar sound which are influenced by palatal sound because the front part of the tongue move from alveolar ridge to the near of soft palate. They are [ʃ], [ʒ], [tʃ], and [dʒ].For example: ‘show’ /ʃəʊ/,

‘vision’ /vɪʒn/, ‘chuck’ /ʧʌk/, and ‘jump’ /ʤʌmp/.

6. Retroflex or post-alveolar

According to Giegerich,the tip of the tongue is curled back and brought near the roof of the mouth behind the alveolar ridgeto produce retroflex or post- alveolar retroflex consonant. So, the articulators are the tip of the tongue as active articulator and the behind of the alveolar ridge as passive articulator. The retroflex consonant is [r], for example, ‘roar’ /rɔ:r/.

7. Palatal

Palatal is produced by raising the front of the tongue towards the hard palate. So, the articulators are the front of the tongue and the hard palate. The front of the tongue acts as active articulator, and the hard palate as passive articulator. The palatal consonant is [j]. For example: ‘you’ /ju/.

8. Velars

Velars are produced by raising the back of the tongue to the soft palate or velum. So, the articulators are the back of the tongue and the soft palate. The active articulator is the back of the tongue, and the passive one is the soft palate.

They are [k], [g], [ŋ], and [x]. For example: ‘key’ /ki:/, ‘gag’ /gæg/, ‘wing’/wɪŋ/, and ‘loch’ /lɒx/.

9. Glottal

Although classified as a consonant, there is no airflow restriction in pronouncing [h]. It’s from the flow of air through the open glottis. The tongue and the lips are usually in the position for the production of the following vowel. If the air is stopped completely in the glottis by tightly closed vocal cords, the sound produced is glottal stop [Ɂ]. The glottal stop also may occur in colloquial speech at the end of the words like don’t, won’t, and can’t. In some American dialects it regularly replaces the ttsound in the words like bottle or glottal (Fromkin and

Rodman, 2003: 243). So, the glottal stop is not the standard English phonology but an allophone or sound variation of [h] which appears in some dialects.

Another example: ‘hang’ /hæŋ/ and ‘bottle’ [bʌɁl].

II. Manners of Articulation

For linguists, it is not enough to describe the consonants by their place of articulation. So, the function of manners of articulation is to distinguish the sounds in each class from each other by the flow of the airstream that may be

blocked or partially blocked; the vocal cord may vibrate or not vibrate (Fromkin and Rodman, 2003: 244).

1. Voiced and Voiceless Sounds

Voiced sounds are produced if the vocal cords are together, and the airstream forces its way through and causes them to vibrate. Simply, the vocal chords are vibrating when produce voiced sounds. The nasal and approximant are also voiced sound. They are [b, d, g, z, ð, v, ʒ, dʒ, m, n, ŋ, l, r, w, j].

Voiceless sounds are the opposite of voiced that the vocal cords are apart during airflow, the air flows freely through the glottis and supraglottal cavities

(the parts of the vocal tract above the glottis). Simply, the vocal cords are not vibrating when produce voiceless sounds. So, the glottal sound is absolutely voiceless. They are [p, t, k, s, θ, f, ʃ, tʃ, h, x, Ɂ, ʍ].

The voiced and voiceless consonants are contrast each other while voiced vibrate the vocal chords and voiceless not. In terms of binary feature, we may distinguish them as [+voiced] for voiced sound, and [-voiced] for voiceless sound.

2. Nasal and oral sounds

Oral sounds are produced with the velum up, blocking the air from escaping through the nose, since the air can escape only through the oral cavity.

Simply, the airstream only can flow from mouth in producing oral sounds. Most sounds in all language are oral sounds including vowels except nasals.

Nasal sounds are produced with the velum not in its raised position, allowing the air escapes through both nose and the mouth. Simply, the airstream only can flow through the nose when producing nasal sounds. So, you can not produce nasal sound clearly if you close your nose when producing nasal sounds.

They are [m], [n], and [ŋ]. For example: ‘mum’ /mʌm/, ‘nun’ /nʌn/, and ‘tank’

/tæŋk/. In binary features, we may distinguished nasal sounds as [+nasal] and oral sounds as [- nasal]

3. Stops and continuant

In producing stop consonants, the airstream may be completely stopped or just partially obstructed. Sounds that are stopped completely in the oral cavity for a brief period are called stops. Sounds in which there is no stoppage in the oral tract are continuants. Oral stops are also called plosive because the air that is blocked in the mouth ‘explodes’ when the closure is released (Fromkin and

Rodman, 2003: 247). They are [pbmtdnkgŋtʃdʒ Ɂ]. They are pronounced in the words like ‘pump’ /pʌmp/, ‘bomb’ /bɒm/, ‘mom’ /mɒm/, ‘tug’ /tʌg/, ‘dig’ /dɪg/,

‘nun’ /nʌn/, ‘keen’ /ki:n/, ‘gin’ /gɪn/, and ‘sing’ /sɪŋ/.

The opposite of stop is continuant. In the continuant, the airstream is not totally blocked in the oral cavity. It can escape continuously through the mouth.

So, the continuant consonants are fricative, liquid, and glide because when producing these consonant the airstream can pass and flow through the mouth.

The nasals are not continuant because the velum blocks the airstream and allow the airstream escape through the nose, not from the mouth. So, we may

distinguish the stop sounds as [+stop] or [-continuant], and continuant sounds as [- stop] or [+continuant].

4. Fricative

In the production of some continuants, the airflow is so severely obstructed that is causes friction, and the sounds are therefore called fricative

(Fromkin and Rodman, 2003: 248). They are [f], [v], [θ], [ð], [s], [z], [ʃ], [ʒ], and

[h]. Fricative sounds appear in the words such as ‘fan’ /fæn/, ‘van’ /væn/, ‘thumb’

/θʌm/, ‘this’ /ðɪs/, ‘sun’ /sʌn/, ‘zip’ /zɪp/, ‘shut’ /ʃʌt/, and ‘vision’ /vɪʒn/. So, in simple binary features, the fricative consonants marked as [+fricative]

5. Affricates

Affricates sounds are produced by a stop closure followed immediately by a gradual release of the closure that produces an effect characteristic of a fricative. So, an affricate is a combination of stop and fricative. They are [tʃ] and

[dʒ]. It presents in the words like ‘church’ /tʃɜ:tʃ/ and ‘judge’ /dʒʌdʒ/. So, the air stream is first interrupted, and then allowed to exit through a narrow channel.

Since air does not escape continuously during the articulation, [ʧ] and [ʤ] are counted as [- continuant]. Like [s] and [z], [ʧ] and [ʤ] are also [+ sibilant].

6. Liquids

In the production of liquids, there is some obstruction of the airstream in the mouth, but not enough to cause any real constriction or friction. They are [l] and [r], as apprears in the words like ‘lady’ /leɪdɪ/ and ‘run’ /rʌn/. In manner of

articulation, both liquids and glides are categorized as approximant because none

of these has radical obstruction of the airstream.

7. Glides

Glides are produced with little or no obstruction of the airstream in the

mouth. In articulating [j] or [w], the tongue moves rapidly in gliding fashion

either toward or away from neighboring vowel. Glides are transitional sounds that

are sometimes called semivowel (Fromkin and Rodman, 2003: 250). They are [j]

and [w]. Glides are produced in words such as ‘you’ /ju/ and ‘way’ /weɪ/. Liquids

and glides are [+approximant].

These descriptions of English consonants in the place and manner of

articulation can be simplified into a chart of English consonant inventory. The first row in

the above of the chart is the member of the place of articulation. The first left column is

the member of the manner of articulation.

Table 2.1 English consonant manners and places of articulation

Place of articulation palato- bilabial labiodental dental alveolar palatal velar glottal Manner alveolar of articulation vs p t k Ɂ stop vd b d g vs f θ s ʃ h fricative vd v ð z ʒ vs ʧ affricates vd ʤ liquid vd l, r glides vd j w nasal vd m n ŋ vs: voiceless vd: voiced

2. The description of Sundanese Phonetics

Because of the laciness of exact source, the Sundanese phonology theory will be compiled from two different sources. They are The Phonology and

Morphology of the Sundanese Language, a doctoral dissertation of philosophy in

University of Michiganwritten by Wayland Bryce Van Syoc in 1959. Based on

Syoc, the Sundanese phonetic system only had eighteen (18) consonants, they are

[p, t, k, b, d, g, m, n, ñ, ŋ, s, h, Ɂ, č, ǰ, l, r, w, j] (Syoc, 1959: 50).

a. Sundanese Consonants

Similar to English consonants, the Sundanese consonants are defined into two classes. They are place of articulation and manner of articulation. Each class member has similar definition with English place and manner of articulation. So it is skipped and just brings the example only except several Sundanese place and manner of articulation.

I. Sundanese Place of Articulation

According to Syoc, the Sundanese place of articulation is divided into six categories. They are bilabial, alveolar, alveo-palatal, palatal, velar, and glottal.

The Sundanese bilabial has three consonants, they are [p, b, and m], for example:

‘pupus’ /pupus/ (dead), ‘budak’ /budak/ (child), ‘mamang’ /mamaŋ/ (uncle).

Sundanese has six alveolar sounds; they are [s, l, r, t, d, n]. For example: ‘saha’

/sahaʔ/ (who), ‘loba’ /lobaʔ/ (many/much), ‘raray’ /raray/ (hair), ‘tiis’ /tiʔis/

(cold), ‘duka’ /dukaʔ/ (I don’t know), ‘naha’ /nahaʔ/ (why).

Syoc divided Sundanese palatal into two class, alveo-palatal and palatal.

Alveo-palatal is a sound which is produced by rising up the tongue blade toward the alveo-palatal region, an area in hard palate which is near from alveolar (Syoc,

1959: 50-4). The definition of Sundanese palatal is similar to English palatal. The

[tš] and [ǆ] are clearly produced in alveo-palatal region.

Besides, both [ñ] and [j] are palatals which are produced by rising up the tongue blade toward the palatal area. The Sundanese [j] has equal manner with

English [j]. Furthermore, the Sundanese [ñ] has two elements, [n] and [j] which slightly move from [n] to [j] which are dominantly occurred in palatal area. So, the Sundanese [ñ] and [j] are palatal. The example for them are ‘caang’ /tšaʔaŋ/

(bright), ‘jajaka’ /ǆaǆakaʔ/ (unmarried young men), ‘nyaho’ /ñahoʔ/ (know), and ‘hayu’ /hajuʔ/ (c’mon).

In the velar consonant, they are [k, g, ŋ]. For example: ‘kalér’ /kaler/

(east), ‘ngojay’ /ŋoǆay/ (swim). Syoc defines the Sundanese semivowel [w] as a vocoid articulated by optional lips rounding accompanied by a rising of the root

(back) of the tongue toward the velum (Syoc, 1956; p 53). So, the semivowel [w] is also classified as the velar sound.

The Sundanese glottal is [Ɂ] and [h]. This glottal stop [Ɂ] in Sundanese always occurs before, after, and between any vowels. So, the feature of [Ɂ] and [h] is contrast in stop and continuant where [Ɂ] is [+stop] or [- continuant], and [h] is

[+continuant] or [- stop]. The example to contrast them is ‘hudang’ /hudaŋ/ (to wake up), and ‘udang’ /Ɂudang/ (shrimp).

II. Sundanese Manner of articulation

According to Sudaryat, Sundanese consonants are classified into seven classes of manner of articulation. The Sundanese voiced consonants are [b, d, g,

ǆ, l, r, m, n, ñ, ŋ, w, y].For example: ‘beubeur’ /bɨbɨr/ (belt), ‘deupa’ /dɨpaɁ/

(measurement), ‘geget’ /gəgət/ (small animal), ‘jeung’ /ǆɨŋ/ (and), ‘loba’ /lobaɁ/

(many/much), ‘rupa’ /rupaɁ/ (kind), ‘amis’ /amɪs/ (good personality), ‘nempo’

/nəmpo/ (to see), ‘nyandak’ /ñandak/ (to take), ‘ngarang’ /ŋaraŋ/ (to write an essay), ‘wewengkon’ /wəwəŋkon/ (place), ‘yakin’ /jakɪn/ (time). The feature for voiced consonant is [+ voiced].

The Sundanese voiceless consonants are [p,t,k,tš,ʔ,s,h]. For example:

‘parang’ /paraŋ/ (a kind of weapon), ‘tari’ /tarɪ/ (to dance), ‘kecap’ /kətšap(word),

‘curug’ /tšurug/ (waterfall), ‘ieu’ /ɁɪɨɁ/ (this), ‘sanes’ (do/does not), ‘hoream’

(lazy). The feature for voiceless consonant is [-voiced].

According to Sudaryat, the Sundanese stop consonants are [p, t, k, b, d, g, tš, ǆ, ʔ]. In manner of articulation, the [tš] and [ǆ] is share same feature with [ʧ] and [ʤ]. They are all stop because the airstream is blocked in a moment. The feature for stop consonant is [+ stop] or [- continuant].

The Sundanese continuant consonants are [s, h, l, m, n, ñ, ŋ, w, j]. Similar to English continuant consonants description, it is produced by allowing the airstream pass through both mouth and nose. So, nasal sounds are classified as continuant.

Sundanese consonants have four nasal sounds; they are [m, n, ñ, ŋ]. The example of Sundanese nasal sounds are ‘mamang’ /mamaŋ/ (an uncle), ‘neneng’

/neneŋ/ (a girl), ‘nyaeta’ /ñaɁeta/ (that is/are), and ‘ngeureut’ /ŋɨrɨt/ (to cut).

Another consonants are oral. There are only two fricative consonants in

Sundanese, [s] and [h]. For example: ‘sora’ /sora/ (sound), and ‘hawa’ /hawa/

(air).

Actually, Sundanese has affricates sounds, but they are somewhat different to English affricates in some way, especially in the place of articulation that English affricates are palato-alveolar than Sundanese affricates are alveo- palatal. They are [tš] and [ǆ], which have equal manner with English affricates

[ʧ] and [ʤ], but different to places of articulation with less lip protrusion and less vocal fold vibration. They are also not found in word final position, thus, causing a problem for an Indonesian in learning English words (Dardjowidjojo, 2009: 44).

Following Dardjowidjojo’s statement, the Sundanese consonants [tš] and [ǆ] have two additional features which distinguish from English affricates, that is,

[less lip protusion] and [less voiced]. Both [tš] and [ǆ] are [+less lip protusion].

But contrast in [less voiced] features that [tš] is [- less voiced] and [ǆ] is [+ less voices].

The liquid consonants in sundanese are [l][r]. For example: ‘letah’ /letah/

(tongue);‘rarangken’ /raraŋken/ (a series of something).Thus, the glides consonants in sundanese are [w, j, ñ].

By the description of Sundanese phonetic above, the Sundanese consonant features by its manners and places of articulation is listed at a table below.

Table 2.2 Sundanese consonants manners and places of articulation

Place of articulation Manner bilabial alveolar Alveo-palatal palatal velar glottal of articulation vs p t k Ɂ stop vd b d g vs s h fricative vd vs tš affricates vd ǆ liquid vd l, r glides vd ñ, j w nasal vd m n ñ ŋ vs: voiceless vd: voiced

3. Theory of Consonant Features

All consonants and vowels could be classified by using the feature as devices for sounds. All of the features we talk about the binary system. So,

[+labial] contrast with [-labial], for example, and every relevant sound has the value of + or – for each feature (Napoli, 1996; 11). It helps us to distinguish each consonant to other consonant. According to Giegerich, there are nine classifications of consonant features.

a. Sonorants:

A sonorant is a sound whose phonetic content is predominantly made up by the sound waves associated with voicing (Giegerich, 1991; 93). The value for

[+son] cover up the approximants and nasal stop. They are [m n ŋ ñ l r w j]. The fricatives and oral stops are [-son]. So, the [-son] consonants are [p t k b d g f v s z

θ ð ʃ ʒ ʧ ʤ].

b. Continuants:

A continuant is a sound during whose production the airstream is not blocked in the oral cavity (Giegerich, 1991; 93). The value of [+cont] covers approximants and fricatives. They are [l r w j ñ f v s z ʃ ʒ θ ð]. The opposite is true, the [-cont] covers nasal and oral stop (that is, ‘stops’ in general) are [-cont].

The affricate could be classified as [-cont], notice that phonetically they are slowly released stops. So, the [-cont] are [m n ŋ p t k b d g ʧ ʤ tš ǆ].

c. Consonantals

A consonantal is a sound produced with a radical obstruction in the vocal tract. All consonants except certain semivowel (liquid approximants) are [+cons], and all vowels are [-cons] (Giegerich, 1991: 94). Unlike semivowel and vowel, of course, liquids are clearly consonantal in phonetic terms. Their production involves a degree of obstruction in the vocal tract that is clearly audible such as radical narrowing in the case of [r], and central contact in the case of [l].

d. Anteriors

Anterior sounds are produced with an obstruction that is located in front of palato-alveolar region of the mouth; non-anterior sounds are produced without such an obstruction (Giegerich, 1991: 116). In terms of English phonetics

description consonants, [+ant] cover the term for bilabial, labiodental, dental, and alveolar. They are [p b m f v θ ð t d n s z l w]. The post-alveolar category (with [r] as the only member) is classified as [-ant], together with palato-alveolar, alveo- palatal, palatal, velar, and glottal sounds. They are [ʃ ʒ ʧ ʤ j k g ŋ h tš ǆ].

e. Coronals

Coronal sounds are produced with the blade of the tongue raised above its neutral position; non-coronal sounds are produced with the blade of the tongue in the neutral position (Giegerich, 1991; 116). The consonants whose [+cor] cover alveolars, palatals, affricates and liquid (Fromkin and Rodman, 2003: 258). They are [t d n l ʧ ʤ tš ʃ ʒ s z ǆ]. The [-cor] segments are bilabials, dentals, labiodentals, velar, glottal, and glides. They are [p b m f v ñ j k g ŋ w].

f. Stridents

Strident sounds are marked acoustically by greater noisiness than their nonstrident counterparts are (Giegerich, 1991; 118). The term refers to the difference in the amount of noise caused by different fricative articulations. The

[+strid] covers up the class of labiodentals, alveolar, alveo-palatal, and palato- alveolar. They are[f v s z ʃ ʒ ʧ ʤ tš ǆ]. Because of its [less lip protrusion] and

[less voiced] features, Sundanese affricates [tš] and [ǆ] are not categorized as strident. So, both [tš] and [ǆ] are [-strident]

g. Voices

A voiced sound is produced with a glottal setting consistent with vocal- fold vibration; a voiceless sound is produced with a glottal setting inconsistent with vocal-fold vibration (Giegerich, 1991; 122). A phoneme specified as

[+voice], then is potentially voiced. It may be fully voiced or to a greater or a lesser extent devoiced depending on its context. They are [b d g m n ŋ ñ l r w j v z

ð ʒ ʤ ǆ]. While a [-voiced] phoneme is voiceless in all contexts. They are [p t k f

θ s ʃ ʧ tš].

h. Nasals

Nasal sounds are produced with a lowered velum, which allows the airstream to escape through the nose (Giegerich, 1991; 124). Non-nasal sounds are produced with a raised velum, so that the airstream can only escape through the mouth. They are [m n ŋ ñ].

i. Laterals

Lateral sounds are produced by lowering the mid section of the tongue at one or both sides, thereby allowing the air to flow out of the mouth in the vicinity of the molar teeth; in non lateral sound no such side passage is open (Giegerich,

1991; 125). English has only one lateral phoneme, which is [l].

These Giegerich’s definitions of consonant features are not satisfied enough. They need an exact feature to distinguish between stop, fricative, and

affricates. According to Carr, there are three more consonant features to solve this problem. They are syllabic, obstruent, and delayed release (Carr, 1991; 55-57).

j. Syllabics

Syllabic or syll for short, is used to distinguish between consonant and vowel. The value of [+syll] is all vowel, and [-syll] is all consonantexcept certain semivowel (glide approximants). This means that all approximant are [-cons, - syll]. So, syllabic is the opposite of consonantal.

k. Obstruents

The term of obstruent refers to the properties of the wave form of the acoustic energy generated in producing consonant. It covers the class of oral stops, affricates, and fricatives. They are [p t k b d g ʧ ʤ s z f v θ ð]. Otherwise, the obstruent is the opposite of the sonorant while [-obs] is [+son]. They are nasal stop [m n ŋ] and approximant [l r w j ñ].

l. Delayed release

The term of delayed release refers to the consonants which are phonetically slowly released stop. Simply talked, the first sound is stop then continued by fricative sound. The class which has value as [+delrel] is only affricates, here, [ʧ] and [ʤ]. To see more the English and Sundanese feature specification at the table below:

Table 2.3 English Consonants Features

m n ŋ p t k b d g f θ s ʃ ʧ h v ð z ʤ ʒ l r w j [cons] + + + + + + + + + + + + + + + + + + + + + + - - [syll] ------[son] + + + ------+ + + + [obs] - - - + + + + + + + + + + + + + + + + + - - - - [cont] ------+ + + + - + + + + - + + + + + [ant] + + - + + - + + - + + + - - - + + + - - + - - - [del rel] ------+ - - - - + - - - - - [cor] - + - - + - - + - - + + + + - + + + + - + - - - [strid] ------+ - + + + - + - + + + - - - - [voice] + + + - - - + + + ------+ + + + + + + + + [nasal] + + + ------[lateral] ------+ - - -

Table 2.4 Sundanese Consonant Features

m n ŋ ñ p t k b d g s h tš ǆ Ɂ l r w j [cons] + + + + + + + + + + + + + + + + + - - [syll] ------[son] + + + + ------+ + + + [obs] - - - - + + + + + + + + + + + - - - - [cont] - - - + ------+ + - - - + + + + [ant] + + - - + + - + + - + - - - - + - - - [del rel] ------+ + - - - - - [cor] - + - + - + - - + - + - + + - + - - - [strid] ------+ ------[voice] + + + + - - - + + + - - - + - + + + + [nasal] + + + + ------[lateral] ------+ - - - [less lip protrusion] ------+ + - - - - - [less voiced] ------+ - - - - -

4. Theory of phonological rules

According to Giegerich, the form of phonological rules is by now

familiar: ‘A is realized as B in the context of C’. The phrase ‘in the context of C’

is present only in allophonic rules, which are context-sensitive; the phonetic

implementation rules mentioned in the preceding section are context free and

merely state ‘A is realised as B’ (Giegerich, 1991; 238). Simply said, A is the

input of the rule, B is the output, and C is the limited term and condition which is cause A become B. See the example below:

V → [+ nasal] ∕ ______[+ nasal] (C) $

The example above is the rule of nasalization. The input is any vowel and the output is the features of the vowel which is [- nasal] become [+ nasal]. The underscore ‘_____’ is used to denote the position of the segment to be changed relative to the conditioning environment. The dollar sign ‘$’ is indicate the syllable boundary. The conditioning environment which is applied (sign with mark slash ‘/’) is ‘before a nasal consonant in the same syllable’. So, the nasalization rule can be read in words: a vowel becomes nasalization in the environment before a nasal segment, possibly followed by a consonant, in the same syllable. For example: ‘bomb’ /bam/ → [bãm].

a. Feature changing rules

According to Fromkin and Rodman, The function of feature changing rules is to change the value of phonemic features (Fromkin, 2003; 305). It means that is when any feature changes in a sound segment; it can be categorized as the feature changing rules. The nasalization rule is classified as assimilation rules, a rules in which a segment sound become more similar to another segment, can be the example of the feature changing rules where the feature of a vowel [- nasal] become [+ nasal]. The dissimilation rules, a rules in which a segment sound become less similar to another segment, also can be the example of the feature changing rules where the feature of a fricative [- stop] become [+ stop]. For

example, the words fifth and sixth come to be pronounced as if they spelled fift and sixt. Again, the devoicing rules of voiced obstruent consonant such as /b d g ð z v/, a rules in which a voiced segment sound become partially or even completely devoiced, also can be the example of the feature changing rules where [+ voice] obstruent consonant become [-voiced] obstruent consonant.

5. Theory of Second Languange Acquisition

In the term second language acquisition, or L2 acquisition, generally refers to the acquisition of a second language by someone (adult or child) who has already acquired a first language (L1). Otherwise, bilingual language acquisition refers to the (more or less) simultaneous acquisition of two languages beginning in infancy (or before the age of three years) (Fromkin, 2003; 374). In the L2 acquisition, the adult or child learners have a basic knowledge of L1 or mother tongue language when they learn the second language. Thus, in the bilingual language acquisition, the learners are always children who learn the two languages at the same time.

Adult second language learners (L2ers) do not often achieve nativelike grammatical competence in the L2, especially with respect to pronunciation. They generally have an accent and they may make syntactic or morphological error that are unlike the errors of children acquiring their first language (L1ers). Unlike L1 acquisition, which is uniformly successful across children and languages, adults vary considerably in their ability to acquire an L2 completely. Some people are very talented language learners. Other people are hopeless. Most people fall

somewhere in the middle. Success may depend on a range of factors, including age, talent, motivation, and whether you are in a country where the language is spoken or sitting in a classroom five morning a week with no further contact of native speakers.

6. Theory of interference

According to Michael Swan and Bernard Smith, the term ‘interference’ or ‘transfer’ mistake categorized as an error (Swan, 2001: xi). In the scope of pronunciation, the interference might happen if the mother tongue has no close equivalent for a feature. If they have, there will no pronouncing problem. So, they carry over their mother tongue pattern to English pronunciation. Here, in the case of Sundanese pronunciation, they do not have complete sound features of fricative compare with English such as [f] and [v]. So, they will ‘transfer’ the sound to the nearest feature that is [p] because of similar labial movement. It assumed that they will pronounce ‘fall’ like ‘paul’.

According to Ellis, the term of interference or negative transfer result from in the cases where the target language differed from the L1. In cases where the pattern of the L1 and the target language (L2) were similar, positive transfer will occur (Ellis, 2003; 29). It means that if there are more differences between L1 and L2, it becomes more difficult to be learnt. Otherwise, if there are more similarities between them, it becomes easier to be learnt. The theory of interference is the basic knowledge of the Contrastive Analysis (CA).

CA is a linguistic enterprise aimed at producing inverted (i.e.contrastive not comparative) two valued typologies (a CA is always concerned with a pair of languages), and founded on the assumption that languages can be compared

(James, 1980; 3). It means that a CA is more interested in differences between languages than in their likenesses. Here, this study focuses on English consonant mispronounced by Sundanese native speaker. So, the research begins with phonological CA before continue with Error Analysis (EA) to find out the consonant differences which cause the mispronunciation.

7. Theory of Error Analysis

Learners make errors in both comprehension and production. An example of a comprehension error is when a learner misunderstand the sentence

‘Pass me the paper’ as ‘Pass me the pepper’, because of an inability to discriminate the sound /eɪ/ and /e/ (Ellis, 2003; 47). Whereas, speech errors serve as the primary data for the construction models of language production (Crookes,

1991 in Ellis, 2003; 71).

The study of error is carried out by means of Error Analysis (EA). In the

1970’s, EA supplanted Contrastive Analysis (CA), which sought to predict the errors that learners make by identifying the linguistic differences between the L1 and target language or TL (Ellis, 2003; 47). So, EA is a development from CA which analyzing the learners error from the linguistics differences between the L1 and TL.

Corder (1974) in Ellis (2003; 48) suggest the following steps in EA research:

1. Collection of a sample of learner language

In the collecting the samples, there are three board types of EA according to the size of the samples. A massive sample involves collecting several samples of language use from a large number of learners in order to compile a comprehensive list of errors, representative of the entire population. A specific sample consist of one sample of language use collected from a limited number of learners, while an incidental sample involves only one sample of language use produced by a single learner (Ellis, 2003; 49). This study is used the second type of EA base on the size of the samples, consider the capacity and the possibility of this study.

2. Identification of errors

An error can be defined as a deviation from the norms of the target languages (Ellis, 2003: 51). This identification of error is request the specific context of which target language is used as the norms. Besides, it need to distinguish between error and mistake.An error (in this technical sense) takes place when the deviation arises as the result of lack of knowledge. It represents a lack of competence (the knowledge of the speaker-hearer has of what constitutes appropriate as well as correct language behaviour in relation to particular communicative goals (see Hymes, 1971a; Canale and Swain, 1980; in Ellis, 2003,

13). A mistake occurs when learners fail to perform their competence. That is, it is

the result of processing problems that prevent the learners from accessing their knowledge of a target language rule, and cause them to fall back on some alternative, non-standard rule that they find easier to access.

3. Description of errors

The description of learner errors involves a comparison of the learner’s idiosyncratic utterances with a reconstruction of those utterances in the target language (Ellis, 2003; 54). It means that it needs to identify the sources of the errors or attention to the surface properties of the learner’s utterances.

Corder (1974) also preserves the framework for describing errors by distinguished three types of errors according to their systematicity (Ellis, 2003;

56):

a. Presystematic errors occur when the learner is unaware of the existence of a

particular rule in the target language. These are random.

b. Systematic errors occur when the learner has discovered a rule but it is the

wrong one.

c. Postsystematic errors occur when the learner knows the correct target

language rule but uses it inconsistently. That is, make a mistake.

4. Explanation of errors

Explanation is concerned with establishing the source of the error, i.e. accounting for why it was made. This stage is the most important for SLA

(Second Language Acquisition) research as it involves an attempt to establish the

process responsible for L2 acquisition (Ellis, 2003; 57). In this step, the source of the error is being explained as the result of L2 acquisition.

5. Evaluation of errors

According to Ellis, error evaluation involves a consideration of the effect that the errors have on the person(s) addressed (Ellis, 2003; 63). This effect can be gauged either in the terms of the addressee’s comprehension of the learner’s meaning or in terms of the addressee’s affective response to the errors. On the other way, the error evaluation analyzes how far the errors influence the learners, both native speakers (NS) and non-native speakers (NNS).

Burt (1975; in Ellis, 2003: 63) divided the errors into two class based on the evaluation. They are global and local errors. Global errors are errors that affect overall sentence organization. Examples are wrong word order, missing and wrongly placed sentence connectors, and syntactic overgeneralizations. Local errors are errors that affect single elements in a sentence (for example, errors in morphology of grammatical functors). It clear that global errors have broader effect errors which consist of lexical and grammatical errors than local which is only in morphological errors, for example.

C. Theoretical Framework

The English and Sundanese phonetics theory are functioned as the basic foundation of this study. That is, to describe the English and Sundanese consonant inventory through the manners and places of articulation. Then, the theory of

consonants features is also used for identify the English and Sundanese consonants features.

The theories of second language acquisition and interference are used to answer the first objective, to find out the similarities and differences between

Sundanese and English phonetics. Moreover, it focuses on the phonetic differences which are caused the consonants mispronunciation. Thus, the theory of error analysis is used to identify the Sundanese errors when they speak English based on phonetics differences between English and Sundanese as the second objectives. Later, the theory of phonological rules is applied for analyzed the

Sundanese errors in pronouncing English consonants as formulated as the third objective.

CHAPTER III METHODOLOGY

A. Object of The Study

The object of the study is the English consonants. To support this study, the form of data to be collected and analyzed was thesound recorded data spoken by Sundanese which contains the predicted mispronunciation of English words from phonetics differences. The predicted mispronunciation of English words were listed based on the English sounds which are absent in Sundanese sound.

B. Method of The Study

1. Data collection

This study was a field research study to collect the speechsound recorded data. So,the most important data were the sound recorded data to find the mispronunciation as interference of Sundanese mother tongue speakers in pronouncing English.Besides, the written data from library and other sources were also important to search the supporting theory both English and Sundanese pronunciation and phonetic rules.

2. Data Analysis

There were two methods applied in this research to answer the problems.

The first method was contrastive analysis to search the similarities and differences between the two languages. The second method was the error analysis which analyzes the mispronunciation happened as the result of the language differences.

The analysis had been done step by step. The first step was to contrast the

English and Sundanese consonants to find out the absent English consonants in

Sundanese which cause the pronunciation difficulties. Then, the second step was to list English words which contain the absent English consonants in Sundanese phonetic in initial, medial, and final position of the words. The purpose was as the indicator of the respondent’s ability in pronouncing the several English consonants, and as the guide for respondents in recording the oral data. The third step was to search the twenty educated Sundanese mother tongue respondents which are the senior high school students in Kuningan, West Java. The purpose was to get the valid data, consider that the senior high school curriculum indicated the students have good performance in speaking English. The fourth step was to record their speech and file them. So, it could be heard many times to check the respondents’ pronunciation. They were the main proof of the respondents’ mispronunciation. The fifth step was to analyze the recorded sound data to find the respondents mispronunciation as the result of interference of Sundanese native speaker in pronouncing English words. The last step was to state the interference and to find out the answer of the third objective.

To support the data recording and analysis, some equipment used. The main equipment were a Labsic headset with an integrated microphone and a notebook Compaq 515 as the data mass storage, and the recording software used the Windows Seven 32 bit sound recorder. Then, the secondary equipment used the Cool Edit Pro 2, sound editing software to cut and edit the sound if necessary to make the process easier.

3. Data indicator

In this research, the data indicators were the words that contained the absent English consonants in Sundanese phonetics which were applied in initial, medial, and final position of the words. To make the research more valid, there were two words applied at each position. Except the English consonant [ʒ], it was very rare to find the English words started by [ʒ]. There were only two words found in the dictionary which starts with consonant [ʒ], ‘gigolo’ and

‘genre’.Because of the moral and norm reason, the word indicator for [ʒ] at initial position only applied one word that was ‘genre’. So, there were total fifty three

(53) words applied as the indicator of the research.

Table 3.1 The Data Indicator Words

Phoneme Initial Medial Final phrase /freɪz/ orphan /ɔ:fn/ stuff /stʌf/ [f] false /fɔls/ effort /efət/ off /əf/ vase /veɪz/ avoid /əvɔɪd/ move /mʊv/ [v] vain /veɪn/ over /əʊvə:/ above /əbʌv/ that /ðæt/ other /ɔðə/ breathe /bri:ð/ [ð] them /ðəm/ mother /mʌðə/ teethe /ti:ð/ thumb /θʌmb/ nothing /nʌθɪŋ/ breath /breθ/ [θ] thank /θæŋk/ birthday /bɜ:θdeɪ/ earth /ɜ:θ/ shop /ʃɒp/ mushroom /mʌʃru:m/ wash /wɔʃ/ [ʃ] she /ʃɪ/ ocean /əʊʃn/ push /pʊʃ/ leisure /leɪʒə/ beige /beɪʒ/ [ʒ] genre /ʒʌŋrə/ measure /mi:ʒə/ rouge /ru:ʒ/ cheap /ʧi:p/ capture /kæpʧə/ bench/benʧ/ [ʧ] chip /ʧɪp/ nature /næʧə/ march /mɑ:ʧ/ joke /ʤəʊk/ subject /sʌbʤekt/ page /peɪʤ/ [ʤ] joy /ʤɔɪ/ soldier /səʊʤɪə/ image /ɪmɪʤ/ zoo /zu:/ museum /mjʊzɪəm/ nose /nəʊz/ [z] zone /zəʊn/ result /rɪzʌlt/ as /əz/

These words above were going to be read by the respondents. Therefore, to make the production of English consonant sound naturally, they were arranged randomize.

4. Data Counting

There were nine (9) English consonants that are considered to be a crucial problematic pronunciation for Sundanese respondents. They were[f],

[v],[ð],[θ],[ʃ],[ʒ],[tʃ],[dʒ],and [z]. Each consonant were scored maximum six(6) because there were two (2)words eachgiven in three (3) position;the initial, medial, and final position. There were read by twenty (20) respondents. So, each consonant had a total score one hundred and twenty (120) from the multiplier of six (6) and twenty (20), except the [ʒ], it had maximum score one hudred (100) from the multiplier of five (5) and twenty (20) because there were only five (5) words given in the consonant [ʒ], one word at the initial, and two (2) words at both medial and final positions. It was formulated at below:

∑

The formula above was used to look for the grade of pronunciation accuracy of both each position and total accuracy which total score for each position was forty (40), from the multiplier of the number of the words at each position (2) and the sum of the respondents (20), except the initial position of [ʒ] which was only twenty (20). So, to find out the consonant pronunciation accuracy at each position, the Σ (sigma, means sum) correct score of each position was

divided by the Σ total score, that was forty (40), except the initial position of [ʒ] that was twenty (20). Thus, to find out the average of the consonant pronunciation accuracy, the Σ correct score of all positions (added the initial, medial, and final position’s score) was divided by the Σ total score in a consonant, that was one hundred and twenty (120), except the consonant [ʒ] that was one hundred.

After find out the percentage of the correct pronunciation, it could be stated that if the percentage was more than 50 % (fifty percent), the English consonant sound was able to be pronounced by the Sundanese respondents.

Otherwise, if it less than 50 %, that consonant was failed to be pronounced. For the details, the grade of accuracy was also divided into ten (10) levels, which were:

Table 3.2 The Grade of Pronunciation Accuracy

Range Grade 90% - 100% Excellent 80% - 89.99% Very good 70% - 79.99% Good 60% - 69.99% Above average 50% - 59.99% Average 40% - 49.99% Below average 30% - 39.99% Poor 20% - 29.99% Very poor 10% - 19.99% Bad 0% - 9.99% Worst

The function of the table above was to indicate the respondents’ accuracy grade both at the each position and each consonant. For example the correct score of [θ] at the initial position is 13, medial position is 17, and final position is 28. So the result for /θ/ is:

= 32.5%

= 42.5%

= 70%

= 48.3%

So, for example, the English consonant [θ] is failed to be pronounced by

Sundanese native speakers in the initial and medial position, but able to be produced well in final position. Thus, the grade of the accuracy of the Sundanese speaker in pronouncing the English consonant [θ] in the initial position is poor, medial position is below average, and final position is good.

CHAPTER IV RESEARCH RESULT

A. Data Description

This chapter is divided into three subchapters. This first subchapter clarifies the data to be analyzed. The data collected in the field are recorded respondents speech. There is not any difficulty recording the data in the field. The data that are analyzed in this chapter are the words list in the table 3.1 at the previous chapter. All of the words are successful to be recorded, sixty-three (63) words total.

The research had been done in SMA Negeri 1 Kuningan, Kuningan, West

Java. There were also twenty (20) respondents whom agree to record their speech.

They were students which their scores were above average in the English test selected by the English teacher. So, there were one thousand and sixty (1060) recorded words speech analyzed, from the multiplier of the word lists (53) and the total respondents (20). Here are the list of the words recorded and the respondents.

Table 4.1 English Word Pronounced By The Respondents

Phoneme Initial Medial Final [f] phrase, false orphan, effort stuff, off [v] vase, vain avoid, over move, above [ð] that, them other, mother breathe, teethe [θ] thumb, thank nothing, birthday breath, earth [ʃ] shop, she mushroom, ocean wash, push [ʒ] genre leisure, measure beige, rouge [ʧ] cheap, chip capture, nature bench, march [ʤ] joke, joy subject, soldier page, image [z] zoo, zone museum, result nose, as

The twenty (20) Sundanese respondents weresenior high school students.

The students were dispersed at several last grade classes. There were six (6) students from the class of XII IPA 2, three (3) students of XII IPS 1, four (4) students of XII IPA 3, and seven (7) students of XII IPA 1. Most of them were females, fifteen respondents were female students, four were male students, and one male teacher. This research is not notice the gender differences in speaking

English but the result of English mispronunciation only.

B. Contrastive Analysis of English-Sundanese Phonetics

This second subchapter is going to discuss the contrastive analysis between English and Sundanese phonetics. The aim is to find out the similarities and differences of both languages, as formulated in problem formulation in chapter one. In chapter two, it had been discussed on both English and Sundanese consonants and vowels, and also their features. This research focuses only at the consonants similarities and differences.

1. Place of articulation

a. Similarities

In the context of place of articulation, Both Sundanese and English have bilabial, alveolar, palatal, velar and glottal sound. There are [p, b, m, w] in the bilabial position in both language. In alveolar position, the Sundanese consonants have similar inventory with English consonant except only [z]. Otherwise, in the palatal region, both English and Sundanese have a consonant sound [j] . In velar

position, both of them have [k, g, ŋ w]. Both English and Sundanese phonetic have glottal sound [h Ɂ].

b. Differences

The Sundanese phonetics does not have labiodentals, dental, some alveolar, palato-alveolar sounds. The labiodentals are [f v]. Although the

Sundanese lexicon has some words inventory which consist of labiodentals consonants, they are not the original Sundanese words. More explanation about labiodentals is going to be discussed in the last section of this chapter. The dentals are [θ ð]. An alveolar is [z]. Palato-alveolars fricative and affricates are [ʃ ʒ ʧ ʤ].

But the Sundanese phonetics have alveo-palatal stop [tš ǆ].

2. Manner of articulation

a. Similarities

Generally, both English and Sundanese phonetic have voiced nasal, stop, fricative, and approximant in manner of articulation. Even though their consonant inventories are different each other. Both of English and Sundanese phonetics has

[b, d, g, m, n, ŋ, l, r, w, j] as voiced consonants. The English and Sundanese nasal consonants are [m, n, ŋ]. The English and Sundanese stop consonants are [p, t, k, b, d, g, m, n, ŋ]. Both The English and Sundanese fricative sounds are only two [s, h]. The English and Sundanese approximants are [l, r, w, j].

b. Differences

Both Sundanese and English phonetics have the affricates sound, but differ in some characteristic that are [less lip protrusion] and [less voiced]. They are voiceless [tš] andvoiced [ǆ] contrast with English [ʧ] and [ʤ]. Some English fricative such as [f, v, θ, ð, ʃ, ʒ, z] also absent in Sundanese. Besides, Sundanese have one more nasal sound that is [ñ] which has the element of [n] and [j], written graphemicly as ‘ny’.

So, from these differences based on both manner and place of articulations, it can be stated that there are nine English consonants which are absent in Sundanese phonetics inventory. They are [f, v, θ, ð, ʧ, ʤ ʃ, ʒ, and z].

Based on the CA theory, these nine consonant cause the pronunciation problem

(mispronunciation) for Sundanese native speakers.

C. Phonological Analysis of The English Consonants Mispronunciation

In this part, it will discuss about consonant distribution of predicted

English consonant mispronounced by Sundanese native speakers. As introduced at previous part, an English consonant can transform into more than oneconsonant as the result of respondent’s mother tongue (Sundanese) and second language

(Indonesian) interference. The consonant distribution itself includes the consonant’s features and the predicted phonological rules through the consonant changed process.

1. Labiodental Fricative Mispronunciation

a. English Consonant [f]

After the research has been done in the field, the English consonant [f] was able to be pronounced by the Sundanese speakers. This fact was shown by the total score for [f] reached up to 70%. It means that the respondents’ accuracy to pronounce this sound is good. Only 30% or some respondents from twenty were failed to produce [f]. They pronounced the English consonant [f] as [p] as appears in words like ‘phrase’ /freɪz/ and ‘orphan’/ɔ:fn/ pronounced as [pʰreɪz] and [ɔ:pʰn].

In initial position, it scored 62.5%. It meant that the accuracy of the respondents who pronounced [f] at the initial position was above averagein the words like ‘phrase’ and ‘false’. In medial position scored 57.5%, meant that the accuracy of respondents who pronounced [f] at medial position was averagein the words like ‘orphan’ and ‘effort’. Thus, final position score is 90%, meant that the accuracy of respondents who pronounced [f] at the final position was excellent in the words like ‘stuff’ and ‘off’.

Table 4.2 Mispronunciation of [f]

English Sundanese % of correct Words Total score pronunciation pronunciation pronunciation phrase [freɪz] [pʰreɪz] 62,5% false [fɔ:ls] [pals] orphan [ɔ:fn] [orpʰan] 57,5% 70% effort [efət] [eport] stuff [stʌf] [stʊp] 90% off [ɒf] [op]

Even though English consonant [f] absent in Sundanese consonant’s inventory, it is able to produce by Sundanese students. Itis influenced by

Indonesian words which are derived from Arabic as religion loan words (notice that West Java has strong religious influences of Muslim). For example, ‘fakir’

[fakɪr] (poor people), ‘faedah’[faɁedah] (benefit), ‘mufakat’ [mʊfakat]

(consensus), ‘kafilah’ [kafɪlah] (desert caravan), ‘wakaf’ [wakaf] (religious foundation), and ‘insyaf’ [insjaf] (realize the mistake). That is why most of respondents are able to avoid mispronounce it.

From the table 4.2, we can see that the English consonant [f] ismostly mispronounced in the words which are begun with spelling‘ph’ and in medial position. This spelling pronunciation also can be predicted appears at final position with same case for example ‘morph’ /mɔ:(r)f/. The consonant distribution of [f] into [p] and [pʰ] occurs when the grapheme ‘ph’ appears at any position in the word. That is why the final position of [f] has the highest score. It does not have any word which contains letter ‘ph’ in final position. In conclusion, even though the consonant distribution of [f] and [pʰ] is predictable in a certain condition, it is unable to be applied into phonological rules. In this case, the distribution of [f] to [pʰ] is not the bilabialized process where voiceless fricative turns into voiceless stop bilabial, but pronounce the English word as its spelling or commonly called spelling pronunciation. It is the common mistakes for

Sundanese speaker.

Another mispronunciation occurs at the other words such as ‘false’,

‘effort’, ‘stuff’, and ‘of’. Though the percentage of the mispronunciation is small,

some of the respondents mispronounce labiodentals fricative [f] with bilabial stop

[p]. It is the result of interference of Sundanese phonetics that there is not any fricative labiodentals sound at all. The sound [f] and [p] contrast each other in phonetic features, they are presented at below:

Table 4.3 Feature Contrast of [f] and [p]

[f] [p] + obstruent +obstruent +continuant - continuant + anterior + anterior - coronal - coronal + strident - strident - voiced - voiced

They contrast only at continuant and strident while the features value of

[f] are [+cont] and [+strid], and /p/ are [-cont] and [-strid]. Both of them share many same features value that are voiceless, [-coronal], [+anterior], and

[+obstruent]. Their voicelessness means the vocal cord does not vibrate; coronalessness means the tongue does not rise up; and the obstruent means they need such an acoustic energy when producing [f] and [p]. They also produced in the same region of anterior. That is the answer why the Sundanese replace [f] to

[p]. The table 4.3.1 shows that it occurs at before, between, and after any vowels.

It can be stated the phonological rules:

 /f/ →/p/ ∕ ______V

The rule above can be read as the consonant [f] becomes [p] in the environmentof preceded by vowels. Forexample: ‘false’ [fɔ:ls].

 /f/ →/p/ ∕ ______C

The rule above can be read as the consonant [f] becomes [p] in the environment of preceded by consonants. For example: ‘phrase’ [freɪz].

 /f/ →/p/ ∕ V_____V

The rule above can be read as the consonant [f] becomes [p] in the environment of between vowels. Forexample: ‘effort’ [efət].

 /f/ →/p/ ∕ V_____C

The rule above can be read as the consonant [f] becomes [p] in the environment ofpreceded by consonants and followed by vowels. For example: ‘orphan’ [ɔ:fn]

 /f/ →/p/ ∕ V_____

The rule above can be read as the consonant [f] becomes [p] in the environment of followed by vowels. For example: ‘stuff’ [stʌf].

Then the rules are applied into the words:

phrase false orphan effort stuff off UR /freɪz/ /fɔ:ls/ /ɔ:fn/ /efət/ /stʌf/ /ɒf/ ↓ ↓ ↓ ↓ ↓ ↓ Sundanese interference [pʰreɪz] [pɔ:ls] [ɔ:pʰn] [epət] [stʌp] [ɒp] PR [pʰreɪz] [pɔ:ls] [ɔ:pʰn] [epət] [stʌp] [ɒp]

b. English consonant [v]

English consonant [v] had total score 46.7%. In the other word, the accuracy of the respondents who produced [v] was below average. It meant that most of Sundanese were failed to produce this consonant as appears at the words like ‘vase’ /veɪz/, ‘vain’ /veɪn/, ‘avoid’ /əvɔɪd/, ‘over’ /əʊvə:(r)/, ‘move’ /mu:v/,

and ‘above’ /əbʌv/. They pronounce [v] became voiceless [f] such as [feɪz],

[feɪnt], [əfɔɪd], [ɔfə(r)], [mʊf], and [əbʌf].

In initial position it scored 30%, meant thatthe accuracy of the respondents who produced [v] was poor at initial position as appears in the words like ‘vase’ and ‘vain’. The score of medial position was 70%, it meantthat the accuracy of the respondents who produced [v] wasgood at medial position as appears in the words like ‘avoid’ and ‘over’. Thus, the score for final position was

40%, meantthat the accuracy of the respondents who produced [v] wasbelow average at final position which appears in the words like ‘move’ and ‘above’.

Table 4.4 mispronunciation of [v]

% of correct Words English Sundanese Total score pronunciation pronunciation pronunciation vase [veɪz] [fas] 30% vain [veɪn] [fein] avoid [əvɔɪd] [afoid] 70% 46.7% over [əʊvə:] [ofer] move [mʊv] [muf] 40% above [əbʌv] [abof]

The Sundanese phonetics system does not have voiced fricative [v] in its inventory. It will be a problem when they learn English and rise misinterpret and misunderstanding when they speak English. The consonant distribution between

[v] and [f] occur at all position in the words. If we noticed the background of the respondents, it was clear that why this consonants distribution occur.

AlthoughIndonesian loan words had consonant [v] in initial and medial position

derived from English words, but they are pronounced as [f], shown at a table below:

Table 4.5contrasting English and Indonesian consonant [v]

English Indonesian Sundanese variation [veərɪeɪʃn] variasi[farɪasɪ] variasi[farɪasɪ] variable [veərɪəbl] variabel [farɪabel] variabel [farɪabel] evaluation [ɪvӕlʊəʃn] evaluasi [efalʊasɪ] evaluasi [efalʊasɪ] evacuation [ɪvӕkjueɪʃn] evakuasi [efakuasɪ] evakuasi [efakuasɪ] massive [mӕsɪv] masif [masɪf] masif [masɪf] active [ӕktɪv] aktif [aktɪf] aktif [aktɪf]

The table 4.4 showed that the voiced fricative [v] changed into voiceless fricative [f] in all position. This process was called devoicing, meant that a voiced sound turned into voiceless but it did not change other features. It was also followed by grapheme change at final position only, note that Indonesian does not have words followed by [v] at the end position.

So,the consonant [v] was directly converted into [f]. That is why

Sundanese were unable to produce voiced fricative sound. Through phonological process, the consonant distribution of /v/ to /f/ could be stated as follow:

 /v/→/f/ ∕ ______V

The rule above can be read as the consonant [v] becomes [f] in the environment of preceded by vowels. For example: ‘vase’ [veɪz]

 /v/→/f/ ∕ V_____V

The rule above can be read as the consonant [v] becomes [f] in the environment of between vowels. For example: ‘avoid’ [əvɔɪd].

 /v/→/f/ ∕ V_____

The rule above can be read as the consonant [v] becomes [f] in the environment of followed by vowels. For example: ‘move’ [mʊv].

Then, the rules are applied into the words:

vase vein avoid over move above UR /veɪz/ /veɪn/ /əvɔɪd/ /əʊvə:/ /mʊv/ /əbʌv/ ↓ ↓ ↓ ↓ ↓ ↓ Sundanese interference [feɪz] [feɪn] [əfɔɪd] [əʊfə:] [mʊf] [əbʌf] PR [feɪz] [feɪn] [əfɔɪd] [əʊfə:] [mʊf] [əbʌf]

2. Dental Fricative consonants

a. English consonant [ð]

English consonant [ð] is failed to pronounce by Sundanese. The accuracy score of pronouncing [ð] is only 16,7%, means that the accuracy of the respondents who produce [ð] is bad. Most of Sundanese are substitute [ð] with

[d], as appear in words like ‘that’ /ðʌt/, ‘them’ /ðəm/, ’other’ /ɔðə(r)/, ‘mother’

/mʌðə(r)/, ‘breathe’/bri:ð/, and ‘teethe’ /ti:ð/.

Table 4.6 The Mispronunciation of [ð]

English Sundanese % of correct Words Total score pronunciation pronunciation pronunciation that [ðæt] [dət] 15% them [ðəm] [dəm] other [ɔðə] [odə] 27.5% 16.7% mother [mʌðə] [madər] breathe [bri:ð] [brit] 7.5% teethe [ti:ð] [tit]

The score for initial position is 15%, means that the accuracy of the respondents who produce [ð] is badat the initial position in the words like ‘that’ and ‘them’. The medial position score is 27.5%, means that the accuracy of the respondents who produce [ð] is very poor at the middle position in the words like

‘other’ and ‘mother’. Later, the final position score is 7.5%, means that the accuracy of the respondents who produce [ð] is worst at the end position of the words like ‘breathe’ and ‘teethe’.

Table 4.7 features contrast of [t], [d], [ð]

[ð] [d] [t] + interdental + alveolar + alveolar + voiced + voiced - voiced -stop + stop + stop + continuant -continuant -continuant + coronal + coronal + coronal + anterior + anterior + anterior - strident - strident - strident + obstruent + obstruent + obstruent

The table 4.6 shows that Sundanese native speakers mispronounce the

English consonant [ð] with [d] in the initial and medial position of the words and with [t] in final position. The English consonant [ð] sounds similar with [d] for

Sundanese speaker, but not [t]. The consonant feature is applied to find out the answer of this anomaly. The features of [ð] are [+ interdental], [+ voiced], [+ continuant], [+coronal], and [+ anterior]. The features of /d/ are [+ alveolar], [+ voiced], [-continuant], [+ coronal], and [+ anterior]. The features of /t/ are [+ alveolar], [- voiced], [- continuant], [+ coronal], and [+ anterior].

From the table 4.7, we can see that the features of [ð] and [d] are contrast in stop, continuant, and their manner of place. Then, the features of [ð] and [t] are contrast in stop, voiced, continuant, and their manner of place. The [ð] changes into [d] because they share similar features such as [+ obstruent], [- strident], [+ anterior], [+ coronal], and [+ voiced].

The distribution of [ð] and [t] only occurs at the end of the words. Notice that there are very rare Indonesian words which are ended by alveolar voiced stop

/d/. Even there are some words such as ‘abad’ and ‘murtad’, they are often pronounced as [abat] and [murtat]. Again, the influence of Indonesian pronunciation is applied here. Moreover, [ð] and [t] share similar features such as

[+ obstruent], [- strident], [+ anterior], and [+ coronal].

So, the phonological rules can be stated as:

 /ð/→/d/ ∕ ______V

The rule above can be read as the consonant [ð] becomes [d] in the environment of preceded by vowels. For example: ‘that’ [ðæt].

 /ð/→/d/ ∕ V____V

The rule above can be read as the consonant [ð] becomes [d] in the environment of between the vowels. For example: ‘other’ [ɔðə].

 /ð/→/t/ ∕ V_____

The rule above can be read as the consonant [ð] becomes [t] in the environment of followed by vowels. For example: ‘teethe’ [ti:ð]

Then the rules are applied in the words:

that them other mother breathe teethe UR /ðæt/ /ðəm/ /ɔðə/ /mʌðə/ /bri:ð/ /ti:ð/ ↓ ↓ ↓ ↓ ↓ ↓ Sundanese interference [dæt] [dəm] [ɔdə] [mʌdə] [bri:t] [ti:t] PR [dæt] [dəm] [ɔdə] [mʌdə] [bri:t] [ti:t]

b. English consonant [θ]

English consonant [θ] is failed to pronounce by Sundanese. It has only scored 25%. It means that the accuracy of the respondents is very poor. Most of them change the English phoneme [θ] into [t] or aspirative [tʰ] as appears in the words like ‘thumb’ /θʌm/, ‘thank’ /θæŋk/, ‘nothing’ /nʌθɪŋ/, ‘birthday’ /bɜ:θdeɪ/,

‘breath’ /breθ/, and ‘earth’ /ɜ:θ/ pronounced as [tʌm], [tæŋk], [nʌtiŋ], [bɜtdeɪ],

[bret], and [ɜt].

The score of initial position is 25%, means that the accuracy of the respondents is very poor at the initial position of the words like ‘thumb’ and

‘thank’. Then, the medial position score is 20%, means that the accuracy of the respondents is very poor at the medial position in the words such as ‘nothing’ and

‘birthday’. Later, the final position score is 30%, means that the accuracy of the respondents is poor at the final position of the words such as ‘breath’ and ’earth’.

Table 4.8 mispronunciation of [θ]

English Sundanese % of correct Words Total score pronunciation pronunciation pronunciation thumb [θʌm] [tam] 25% thank [θæŋk] [teŋ] nothing [nʌθɪŋ] [notɪng] 20% 25% birthday [bɜ:θdeɪ] [bətdeɪ] breath [breθ] [bret] 30% earth [ɜ:θ] [ə:t]

Because there is no interdental fricative, they change it with alveolar stop

[t].The English consonant [θ] sound similar with [t] for Sundanese. Actually they have distinguished features. The features of [θ] are [+ fricative], [- voiced],

[+interdental], [+ continuant]. The features of [t] are [- fricative], [- voiced],

[+alveolar], [- continuant]. The comparison between them is contrast at the table below:

Table 4.9 features contrast of [t] and [θ]

[θ] [t] + fricative - fricative - voiced - voiced + interdental + alveolar + continuant - continuant + anterior + anterior + coronal + coronal + obstruent + obstruent - strident - strident

The conclusion is the consonant [θ] and [t] is contrast in fricative, stop or continuant, and manner of place. They have similarities in voiceless, anterior, and coronal. The manner of place between [θ] and [t] is very close which [θ] is interdental and [t] is alveolar. The consonants have changed from interdental

(between the teeth) to alveolar (alveolar ridge). So, from this explanation, it can be arranged the rules:

 /θ/→/t/ ∕ ______V

The rule above can be read as the consonant [θ] become [t] in the environment of preceded by vowels. For example: ‘thumb’ [θʌm]

 /θ/→/t/ ∕ V_____V

The rule above can be read as the consonant [θ] become [t] in the environment of preceded between vowels. For example:‘nothing’ [nʌθɪŋ]

 /θ/→/t/ ∕ V_____C

The rule above can be read as the consonant [θ] become [t] in the environment of preceded by consonants and followed by vowel. For example:‘birthday’

[bɜ:θdeɪ]

 /θ/→/t/ ∕ V______

The rule above can be read as the consonant [θ] become [t] in the environment of followed by vowels. For example:‘breath’ [breθ]

Then the rules are applied into the words:

thumb thank nothing birthday breath earth UR /θʌm/ /θæŋk/ /nʌθɪŋ/ /bɜ:θdeɪ/ /breθ/ /ɜ:θ/ ↓ ↓ ↓ ↓ ↓ ↓ Sundanese interference [tʌm] [tænk] [nʌtɪŋ] [bɜ:tdeɪ] [bret] [ɜ:t] PR [tʌm] [tænk] [nʌtɪŋ] [bɜ:tdeɪ] [bret] [ɜ:t]

3. Palato-Alveolar Fricative Consonants

a. English Consonant [ʃ]

Sundanese are able to produce the English consonant [ʃ]. It was shown by the result of the research that reached 70%. It means that the accuracy of the respondents is good. They exchange this consonant to [s] and [tš] as shown at the words ‘shop’ /ʃɔp/, ‘she’ /ʃɪ/, ‘mushroom’ /mʌʃru:m/, ‘ocean’ /əʊʃn/, ‘wash’ /wɔʃ/, and ‘push’ /pʊʃ/.

In the initial position, it result 57.5%, means that the accuracy of the respondents who pronounce [ʃ] is average at the initial position in the words such as ‘shop’ and ‘she’. The score for medial position is 75%, means that the accuracy of the respondents is good at the medial position in the words such as ‘mushroom’ and ‘ocean’. The final position score is 77.5%, means that the accuracy of the respondents is good at the final position in the words such as ‘wash’ and ‘push’.

Table 4.10 Mispronunciation [ʃ]

English Sundanese % of correct Words Total score Pronunciation Pronunciation pronunciation shop [ʃɒp] [sop] 57.5% she [ʃɪ] [sɪ] mushroom [mʌʃru:m] [musru:m] 75% 70% Ocean [əʊʃn] [otšean] wash [wɔʃ] [was] 77.5% push [pʊʃ] [pus]

Although most of respondents does not have problem with this English consonant, but several mistakes which occurs are still interesting to be discussed.

This research found that the English consonant [ʃ] often mispronounced as [s],

except the respondents’ confusion with [tš] as appears in the word like ‘ocean’, but there is not doubt with ‘mushroom’. In the features of consonant, [ʃ] have

[+palatal], [- voiced], [+ fricative], [+ coronal], and [+ strident]. Otherwise, [s] have [+ alveolar], [-voiced], [+ fricative], [+coronal], and [+ strident]. The contrasting table is viewed below:

Table 4.11 features contrast of [ʃ] and[s]

[ʃ] [s] + consonantal + consonantal - voiced -voiced + fricative + fricative + strident + strident + coronal + coronal + continuant + continuant + obstruent + obstruent - anterior + anterior

From the table above, it is clear that [ʃ] and [s] distinguish on the anterior only, means that they are different in place of articulation only. It can be stated that the distribution of [ʃ] and [s] through the process of become alveolar sound when the position of the front of tongue has been changed by moving forward from hard palate to alveolar ridge.

The distribution of [ʃ] and [s] occur in initial and final position. It was indicated by the letter ‘sh’ in the English words preceded by vowels in initial position and followed by vowels in final position. Indeed, there are many English words that contain [ʃ] sound without letter ‘sh’ like ‘asian’ /eɪʃən/, ‘mission’

/mɪʃn/, ‘nation’ /neɪʃn/, ‘luscious’ /lʌʃəs/, ‘election’ /ɪlekʃn/, ‘delicious’ /delɪʃɪəs/, and ‘mention’ /menʃn/. This situation makes the Sundanese speakers confused

when they meet those words. So, the respondents applied the spelling pronunciation rules for any consonant. So, the rules can be stated as:

 /ʃ/→/s/ ∕ ______V

The rule above can be read as the consonant [ʃ] become [s] in the environment of preceded by vowels. For example:‘shop’ [ʃɒp]

 /ʃ/→/s/ ∕ V______C

The rule above can be read as the consonant [ʃ] become [s] in the environment of preceded by consonants and followed by vowels. For example: ‘mushroom’

[mʌʃru:m]

 /ʃ/→/s/ ∕ V______

The rule above can be read as the consonants [ʃ] become [s] in the environment of followed by vowels. For example:‘wash’ [wɔʃ]

Then, the rules are applied in the words:

UR ‘shop’ ‘she’ ‘mushroom’ ‘ocean’ ‘wash’ ‘push /ʃɒp/ /ʃɪ/ /mʌʃru:m/ /əʊʃn/ /wɔʃ/ /pʊʃ/

↓ ↓ ↓ ↓ ↓ ↓

Sundanese interference [sop] [sɪ] [mʌsru:m] [otšean] [wɔs] [pʊs] PR [sop] [sɪ] [mʌsru:m] [otšean] [wɔs] [pʊs]

b. English Consonant [ʒ]

As predicted before, Sundanese get difficulty to produce English consonant [ʒ]. The result is 14.17%, means that the accuracy of the respondents is bad. Most of them mispronounce the English consonant [ʒ] as [g] and spelling pronunciation as appear in the words like ‘genre’ /ʒʌŋre/, ‘leisure’ /leɪʒɜ:/,

‘measure’ /meʒɜ:/, ‘beige’ /beɪʒ/, and ‘rouge’ /ru:ʒ/.

For details, the score of initial position is 0%, means that all respondents from twenty are unable to produce English consonant [ʒ] at the initial position of a word like ‘genre’. The score of medial position is 12.5%, means that the accuracy of the respondents who pronounce the English consonant [ʒ] is bad at the medial position in the words like ‘leisure’ and ‘measure’. Then, the score of final position is 30%, means that the accuracy of the respondents who pronounce the English consonant [ʒ] is poor at the final position of the words such as ‘beige’ and ‘rouge’.

Table 4.12 mispronunciation of [ʒ]

English Sundanese % of correct Total Words pronunciation pronunciation pronunciation score genre [ʒʌŋrə] [genrə] 0% leisure [leɪʒə] [leisur] 12.5% measure [mi:ʒə] [mesur] 14.17% beige [beɪʒ] [beig] 30% rouge [ru:ʒ] [ru:g]

The English consonant [ʒ] seems very difficult to pronounce by

Sundanese. It was shown from its accuracy which is lowest from other consonants. In initial position, the distribution always occurs before vowel, and it is changed to [g]. Then, the distribution of [ʒ] and [s] occurs only in medial

position between the vowels. Thus, in final position there is only the distribution of [ʒ] and [g] after the vowel.

From the table above, the respondents pronounce these words similar with their spelling in all position, or spelling pronunciation. So, it is not valid to apply the features contrasting method to find out the distribution process, but apply the rules of spelling pronunciation:

 /ʒ/→/g/ ∕ ______V

The rule above can be read as the consonant [ʒ] becomes [g] in the environment of preceded by vowels. For example: ‘genre’ [ʒʌŋrə]

 /ʒ/→/s/ ∕ V______V

The rule above can be read as the consonant [ʒ] becomes [s] in the environment of between vowels. For example:‘leisure’ [leɪʒə]

 /ʒ/→/g/ ∕ V______

The rule above can be read as the consonant [ʒ] becomes [g] in the environment of followed by vowels. For example: ‘beige’ [beɪʒ]

Then, the rules is applied in the words:

UR genre leisure measure beige rouge /ʒʌŋrə/ /leɪʒə/ /mi:ʒə/ /beɪʒ/ /ru:ʒ/ ↓ ↓ ↓ ↓ ↓ Applying the spelling pronunciation [genrə] [leisure] [mesur] [beig] [ru:g] rules PR [genrə] [leisure] [mesur] [beig] [ru:g]

4. Palato-Alveolar Affricates Consonants

a. English Consonant [ʧ]

The research result show that the respondents are able to pronounce the

English consonant [ʧ] as shown by result is 58.3%. It means that the accuracy of the respondents is average. The words which are used as the indicator are ‘cheap’

/ʧi:p/, ‘chip’ /ʧɪp/, ‘capture’ /kæpʧə/, ‘nature’ /næʧə/, ‘bench’ /benʧ/, and ‘march’

/mʌrʧ/. Some respondents mispronounced it with [tš] at the initial position and final position, and [t] at the medial position.

The result for initial position is 80%, means that the accuracy of the respondents is very goodat the initial position as appears in the words such as

‘cheap’ and ‘chip’. The medial position result is 17.5%, means that the accuracy of the respondents is bad at the medial position in the words like ‘capture’ and

‘nature’. The score of final position is 77.5%, means that the accuracy of the respondents is good at the final position in the words such as ‘bench’ and ‘march’.

Table 4.13 Mispronunciation of the [ʧ]

English Sundanese % of correct Words Total score pronunciation pronunciation pronunciation cheap [ʧi:p] [tšɪp] 80% chip [ʧɪp] [tšɪp] capture [kæpʧə] [kaptur] 17.5% 58.3% nature [næʧə] [natur] bench [benʧ] [bents] 77.5% march [mɑ:ʧ] [martš]

The result of this research shows that the mispronunciation of English sound segment [ʧ] occurs at all position. In fact, most of respondents are unable to

produce [ʧ] at medial position rather than another position. They pronounced the words as their spelling such as ‘capture’ and ‘nature’. So, it could be predicted they will do same mistake with another words, for example: ‘actual’, ‘feature’, and ‘picture’.

From the table above, it is clear that the distribution of [ʧ] and [tš] occurs in initial and final position. Sundanese respondents change the first and last consonant with Sundanese voiceless alveo-palatal affricates because there is no voiceless palato-alveolar affricate [ʧ] in Sundanese consonants inventories. So, the front of tongue is move forward from soft palate to alveolar ridge, and blocks the airflow.

Table 4.14 features contrast of [ʧ]and [tš]

[ʧ] [tš] + obstruent + obstruent - continuant -continuant - anterior - anterior + delayed release + delayed release + coronal + coronal + strident - strident -voiced - voiced + lips protrusion - lips protrusion

The English [ʧ] and Sundanese [tš] are only contrast in the features of

[strident] and [lip protrusion]. But they share same features of obstruent, continuant, anterior, delayed release, coronal, and voiced. That is the answer of the Sundanese mispronunciation of [ʧ] to [tš]. Though their similarities are in the region of anterior and coronal, but the rising tongue position in producing the palato-alveolar [ʧ] and alveo-palatal [tš] are different. Again, the English consonant [ʧ] is more lips protrusion than Sundanese [tš].

The phonetic rules for the distribution of [ʧ] and [tš] state as:

 /ʧ/→/tš/ ∕ ______V

The rule above can be read as the consonant [ʧ] becomes [tš] in the environment preceded by vowels. For example: cheap [ʧi:p].

 /ʧ/ → /t/ ∕ V_____V

The rule above can be read as the consonant [ʧ] becomes [tš] in the environment between the vowels. For example: ‘nature’ [næʧə].

 /ʧ/ → /t/ ∕ C_____V

The rule above can be read as the consonant [ʧ] becomes [tš] in the environment preceded by the vowels and followed by the consonants. For example: ‘capture’

[kæpʧə].

 /ʧ/ →/tš/ ∕ C______

The rule above can be read as the consonant [ʧ] becomes [tš] in the environment followed by the consonants. For example: ‘bench’ [benʧ]

Then, the rules are applied in the words:

UR ‘cheap’ ‘chip’ ‘capture’ ‘nature’ ‘bench’ ‘march /ʧi:p/ /ʧɪp/ /kæpʧə/ /næʧə/ /benʧ/ /mɑ:ʧ/ ↓ ↓ ↓ ↓ ↓ ↓ Sundanese interference [tšɪp] [tšɪp] [kaptur] [natur] [bentš] [martš] PR [tšɪp] [tšɪp] [kaptur] [natur] [bents] [martš]

b. English Consonant [ʤ]

Then, another English consonant is [ʤ]. The result of this research is

66.7%, means that the accuracy of the respondents is above averageto pronounce the English consonant [ʤ] in the words such as ‘joke’ /ʤəʊk/, ‘joy’ /ʤoɪ/,

‘subject’ /səbʤek/, ‘soldier’ /səʊlʤɪə/, ‘page’ /peɪʤ, and ‘image’ /ɪmɪʤ/ as the indicator. Another respondents are unable to produce it, and change into another consonant such as [ǆ], [d], and [g].

The score for initial position is 85%, means that the accuracy of the respondents is very good at the initial position in the words such as ‘joke’ and

‘joy’. The result of medial position is 52.5%, means that the accuracy of the respondents is average at the medial position in the words like ‘subject’ and

‘soldier’. The score for final position is 62.5%, means that the accuracy of the respondents is above average at the final position as appears in the words such as

‘page’ and ‘image’.

Table 4.15 mispronunciation of [ʤ]

English Sundanese % of correct Words Total score pronunciation pronunciation pronunciation joke [ʤəʊk] [ǆok] 85% joy [ʤɔɪ] [ǆoy] subject [sʌbʤekt] [subǆek] 52,5% 66.7% soldier [səʊʤɪə] [soldier] page [peɪʤ] [peig] 62,5% image [ɪmɪʤ] [imeig]

The Sundanese respondents substitute the English palato-alveolar voiced affricates [ʤ] with alveo-palatal less voiced [ǆ]. It occurs in the words which

content of consonant spelling ‘j’ at the initial and medial position such as ‘joke’,

‘joy’, and ‘subject’. Otherwise, the other words which does not have consonant spelling [j] but pronounced as /ʤ/, apply the spelling pronunciation rules, like in

[soldier], [page], and [image]. The English phoneme [ʤ] is sound similar with

Sundanese phoneme [ǆ], but very contrast in their features.

Table 4.16 features contrast of [ʤ] and [ǆ]

[ʤ] [ǆ] + obstruent + obstruent - continuant - continuant - anterior - anterior + delayed release + delayed release + strident - strident + voiced + less voiced + coronal + coronal + lips protrusion - lips protrusion

From the table 4.16, it shows that they contrast in [strident], [voiced], and

[lips protrusion]. Although Sundanese consonant [ǆ] is an affricate sound, but somehow different with English affricate. The Sundanese [ǆ] is [voiced], but the degree of [voiced] is less than English [ʤ]. Together with [lips protrusion] features which [ǆ] is less protrudes than [ʤ]. They are also different in place of articulation and manner of articulation, which [ʤ] is palato-alveolar and [ǆ] is alveo-palatal consonant. Their places of articulation are very close, and both of them share same feature [+ obstruent], [- cont], [- ant], [+ delrel], and [+ coronal].

Based on their similarities, the English consonant [ʤ] mispronounced with [ǆ] by Sundanese speakers. Another [ʤ] mispronunciation is spelling pronunciation that is pronounced as its spelling. So, we can state the rules of /ʤ/ mispronunciation as follow:

 /ʤ/ → /ǆ/ ∕ ______V

The rule above can be read as the consonant [ʤ] becomes [ǆ] in the environment of preceded by vowels. For example: ‘joke’ [ʤəʊk]

 /ʤ/ → /ǆ/ /d/ ∕ C_____V

The rule above can be read as the consonant [ʤ] becomes [ǆ] in the environment of preceded by vowels and followed by consonants, and becomes [d] as the result of spelling pronunciation. For example:‘subject’ [sʌbʤekt], ‘soldier’ [səʊʤɪə]

 /ʤ/ → /g/ ∕ V______

The rule above can be read as the consonant [ʤ] becomes [g] in the environment of followed by vowels as the result of spelling pronunciation. For example: page[peɪʤ]

Then apply the rules into the words:

UR ‘joke’ ‘joy’ ‘subject’ ‘soldier’ ‘page’ ‘image’ /ʤəʊk/ /ʤɔɪ/ /sʌbʤekt/ /səʊʤɪə/ /peɪʤ/ /ɪmɪʤ/ ↓ ↓ ↓ ↓ ↓ ↓ Sundanese interference [ǆok] [ǆoy] [subǆek] [soldier] [peɪg] [ɪmɪg] PR [ǆok] [ǆoy] [subǆek] [soldier] [peɪg] [ɪmɪg]

5. Alveolar fricative

a. English Consonant [z]

Last English consonant to be analyzed in this research is [z]. The total result for this sound is 17.5%, means that the accuracy of the respondents is badas

indicate by the words such as ‘zoo’ /zu:/, ‘zone’ /zəʊn/, ‘museum’ /mjuzɪəm/,

‘result’ /rɪzʌlt/, ‘nose’ /nəʊz/, and ‘as’ /əz/. Most of Sundanese (83.5%) are change it to another consonant such as [s] and [ǰ].

At initial position, it scores 25%, means that the accuracy of the respondents is very poor at initial position of the words such ‘zoo’ and ‘zone’.

The score of medial position is 15%, means that the accuracy of the respondents is bad at medial position words such as ‘museum’ and ‘result’. Last, the final position score is 12.5%, means that the accuracy of the respondents is bad at final position words such as ‘nose’ and ‘as’.

Table 4.17 mispronunciation of [z]

English Sundanese % of correct Words Total score pronunciation pronunciation pronunciation zoo [zu:] [ǆu] 25% zone [zəʊn] [ǆon] museum [mjʊzɪəm] [musəum] 15% 17.5% result [rɪzʌlt] [result] nose [nəʊz] [nos] 12.5% as [əz] [əs]

In the alveolar fricative region, there is one English sound segment which is unable to produce by Sundanese because its absence in Sundanese phonetics system, that is [z]. The result of this research show that the mispronunciation of

[z] occurs in initial, medial and final position. In initial position, this consonant changes to the alveo-palatal stop voiced [ǆ] before any vowels. But this rule does not applicable at the medial and final position. The respondents apply the spelling pronunciation rules.

Table 4.18features contrast of [z],[ǆ] and [s]

[z] [ǆ] [s] + obstruent + obstruent + obstruent + continuant - continuant + continuant + anterior - anterior + anterior + strident - strident + strident + voiced + less voiced - voiced +coronal + coronal + coronal

From the table 4.18, we can see that the [z] and [ǆ] are contrast in continuant, voiced, anterior, and strident. Besides, they have similarities of features such as obstruent and coronal consonant. It means that they have fewer similarities in place and manner of articulation than its differences. The reason of the mispronunciation of [z] to [ǆ] is the local linguistic and cultural influence.

The Sundanese also speaks Indonesian as second language. Indonesian words have consonant letter ‘z’ in the initial and medial position, and never present at the final position. But it always pronounce as [ǆ] or [s], for example as appears in

‘zaman’ [ǆaman] (age), ‘zirah’ [ǆɪrah] (armored clothes), izin [ɪǆɪn]

(permission), ijazah [ɪǆasah] (lisence). There are also devoicing process which voiced [z] change become voiceless [s], but it seems pronounced same as its grapheme, or called spelling pronunciation.

So we can state the rule of [z] mispronunciation as follow:

/z/ → /ǆ/ ∕ ______V

The rule above can be read as the consonant [z] becomes [ǆ] in the environment preceded by vowels. For example:‘zoo’ [zu:]

/z/ → /s/ ∕ V______V

The rule above can be read as the consonant [z] becomes [s] in the environment between the vowels. For example:‘museum’ [mjʊzɪəm]

/z/ → /s/ ∕ V______

The rule above can be read as the consonant [z] becomes [s] in the environment followed by vowels. For example:‘nose’ [nəʊz]

Then apply the rules into the words:

UR ‘zoo’ ‘zone’ ‘museum’ ‘result’ ‘nose’ ‘as’ /zu:/ /zəʊn /mjʊzɪəm/ /rɪzʌlt/ /nəʊz/ /əz/ ↓ ↓ ↓ ↓ ↓ ↓ Sundanese interference [ǆu] [ǆon] [musəum] [result] [nos] [əs] PR [ǆu] [ǆon] [musəum] [result] [nos] [əs]

D. Summary of the Research

1. Contrastive Analysis of English and Sundanese Consonants

After contrasting between the English and Sundanese consonants system and feature, English have more consonants than Sundanese. The English consonants which are absent in Sundanese are [f, v, θ, ð, ʃ, ʒ, ʧ, ʤ, z]. They are classified as labiodental fricative [f] and [v], dental fricative [θ] and [ð], alveolar fricative [z], palato-alveolar fricative [ʃ] and [ʒ], and palato-alveolar affricates [ʧ] and [ʤ]. These absence will cause the mispronunciation when the Sundanese speaker speaking English.

Both English and Sundanese have consonant [p, t, k, b, d, g, m, n, ŋ, s, l, r, w, j]. They will not cause any problem in pronunciation. The Sundanese consonant inventory have some consonant sound which are absent in English, they are [tš, ǆ, ñ, Ɂ]. ,

2. Learner’s Error In Pronouncing English Consonants

After the research done, the result of Sundanese speaker’s English

mispronunciation varied in its accuracy and consonant change. The accuracy of

English labiodental voiceless fricative [f] pronunciation reached 70% in which

62.5% at initial position, 57.5% at medial position, and highest 90% at final

position. It means that the Sundanese speakers’ pronunciation accuracy is good.

Some of Sundanese speakers mispronounced it into bilabial voiceless stop [p] as

the language interference.It was also the influences of Indonesian phonetics

system.

The accuracy of English labiodental voiced fricative [v] pronunciation

reached 46.7% in which 30 % at initial position, 70% at medial, and 40% at final

position. It means that the Sundanese speakers’ pronunciation accuracy is below

average in rate, but good in medial position. Most of Sundanese speakers

mispronounced it into labiodental voiceless fricative [f] as the language

interference.

The accuracy of English interdental voiceless fricative [θ] pronunciation

reached 25% in which 25% at initial position, 20% at medial position, 30% at

final position. It means that the Sundanese speakers’ pronunciation accuracy is

very poor in rate. Most of Sundanese speakers mispronounced it into alveolar voiceless stop /t/ as the language interference.

The accuracy of English interdental voiced fricative [ð]pronunciation reached 16.67% in which 15% at initial position, 27.5% at medial position, and worst7.5% at final position. It means that the Sundanese speakers’ pronunciation accuracy is bad in rate. Most of Sundanese speakers mispronounced it into alveolar voiced stop [d] as the language interference.

The accuracy of English interdental voiced fricative [ʃ]pronunciation reached 70% in which 57.5% at initial position, 75% at medial position,and

77.5% at final position. It means that the Sundanese speakers’ pronunciation accuracy is good in rate. Most of Sundanese speakers mispronounced it into alveolar voiced stop [s] as the language interference.

The accuracy of English interdental voiced fricative [ʒ]pronunciation reached 14.16% in which 0% at initial position, 12.5% at medial position,and

30% at final position. It means that the Sundanese speakers’ pronunciation accuracy is bad in rate and worst in initial position. Most of Sundanese speakers mispronounced it into alveo-palatal voiced stop /ǆ/ as the language interference.

The accuracy of English palato-alveolarvoicelessaffricate

[ʧ]pronunciation reached 58.33% in which 80% at initial position, 17.5% at medial position, and 77.5% at final position. It means that the Sundanese speakers’ pronunciation accuracy is average in rate but very good in initial

position. Someof Sundanese speakers mispronounced it into alveo-palatal voicelessstop [tš] as the language interference.

The accuracy of English palato-alveolar voiced affricates

[ʤ]pronunciation reached 66.67% in which 85% at initial position, 52.5% at medial position, and 62.5% at final position. It means that the Sundanese speakers’ pronunciation accuracy is above average in rate but very good in initial position. Some Sundanese speakers mispronounced it into alveo-palatal voiced stop [ǆ] as the language interference.

The accuracy of English alveolarvoiced fricative [z]pronunciation is reached 17.5% in which 25% at initial position, 15% at medial position, and12.5% at final position. It means that the Sundanese speakers’ pronunciation accuracy is bad in rate. Most of Sundanese speakers mispronounced it into alveolar voicelessfricative [s] and alveo-palatal voiced stop [ǆ] as the language interference.

CHAPTER V

CONCLUSION

After reading the theory and doing the research, the Sundanese speakers have develop better pronunciation of some English consonants sound shown by the grade of accuracy. Base on the result of contrastive analysis, there are nine (9)

English consonants sound which are predicted to be a pronunciation problem for the Sundanese speaker from its phonetics differences. The result of this research found the same things as predicted before.

The answer of first problem is that the Sundanese phonetic inventory does not have English consonants [f v θ ð ʃ ʒ ʧ ʤ z]. These English consonants will be difficult to be pronounced by the Sundanese speaker. Both English and

Sundanese have similarities with other consonants such as [p t k b d g m n ŋ s h l r w j].

Actually, the Sundanese speakers have a good accuracy in pronouncing

[f] and [ʃ] (70%), above average in pronouncing [ʤ] (66.67%), and average in pronouncing [ʧ] (58.33), the five (5) consonants are scored under 50%. The lowest accuracy is in the pronouncing [ʒ] (14.16%), followed by [ð] (16.67%), continued by [z] (17.5%), then [θ] (25%), and the last is [v] (46.67%).

Most of Sundanese speakers’ mispronunciations are the spelling pronunciation as the language interference of the mother tongue. The example of the spelling pronunciation happen in almost English consonants, one of them is

‘genre’/ʒʌŋrə/is pronounced as [gənrə].The solution to prevent the spelling

pronunciation is checking the pronunciation of the words in the dictionaries like

Oxford and Merriam-Webster dictionaries.

Another mispronunciation is devoicing mispronunciation when a voiced consonant become voiceless consonants. It is also the interference from the

Indonesian pronunciation. Notice that the respondents are also learning Bahasa

Indonesia at school and also use it as the medium language of teaching. The examples of the devoicing mispronunciation are appears in the English consonant

/v/ is ‘vase’ /vəɪz/ is pronounced as [fas].The solution to prevent the devoicing mispronunciation is trying to vibrate the vocal chord when produces the voiced consonants like [v] to differ with voiceless [f].

The last kind of mispronunciation is the mispronunciation of two consonants which are in a close place of articulation. It occurs between the two close places of articulations such as labiodental sound /f/ is pronounced as bilabial

/p/ for example ‘off’ /ɒf/ is pronounced as [ɒp]; interdental /θ/ and /ð/ is pronounced as alveolar sound /t/ and /d/ for example ‘nothing’ /nʌθɪŋ/ is pronounced as [notɪng] and ‘them’ /ðəm/ is pronounced as [dəm]; palato- alveolar /ʧ/ and /ʤ/ is pronounced as alveo-palatal sound /tš/ and /ǆ/ for example

‘cheap’ /ʧi:p/ is pronounced as [tšɪp], and ‘joke’ /ʤəʊk/ is pronounced as

[ǆok]; palato-alveolar /ʃ/ is pronounced as alveolar sound /s/ for example ‘shop’

/ʃɒp/ is pronounced as [sop]. After the research done, it could be concluded that the learning process would be easier if there were more similarities between the two languages. Otherwise, it would be harder if there were more differences between the two languages.

BIBLIOGRAPHY

Alip, Francis Borgias. The Feasibility of Indonesian English.Phenomena Journal vol 8 no. 1. June 2004.

Carr, Philip. Phonology.London: The MacMillan Press LTD, 1993.

Carruthers, Steven W. Pronunciation Difficulties of Japanese Speakers of English. TESL Working Paper Series Vol 4, Issue 2. Fall 2006.

Collins, Beverley and Inger M. Mees.Practical Phonetics and Phonology: A Resource Book for Students. London: Routledge, 2003.

Dulay, Heidi, Marina Burt, and Stephen Krashen.Language Two.New York: Oxford University Press, 1982.

Ellis, Rod. The Study of Second Languange Acquisition. Oxford: Oxford University Press, 2003.

Fasold, Ralph, and Jeff Connor-Linton.An Introduction to Language and Linguistics.New York: Cambridge University Press, 2006.

Fromkin, Victoria, Robert Rodman, and Nina Hyams.An Introduction to Language – seventh edition. Boston: Heinle Thomson Corporation, 2003.

Indriani, M.I. English Pronunciation: The English Speech Sounds Theory and Practice. Jakarta: PT GramediaPustakaUtama, 2005.

James. Contrastive Analysis.Harlow: Addison Wesley Longman Limited, 1980.

Kreidler, Charles W. The Pronunciation of English: a Course Book. Cornwall: Blackwell Publishing, 2004.

Napoli, Donna Jo. Linguistics.New York: Oxford University Press,1996.

Richard, Jack C. Error Analysis: Perspective on Second Language Acquisiton. London: Longman Group Limited, 1974.

Roach, Peter. Phonetics.Oxford: Oxford University Press, 2001.

Swan, Michael and Bernard Smith.Learner English.Cambridge: Cambridge University Press, 2001.

Sudaryat, Yayat. UlikanFonologiBasaSunda. Bandung: JPBD FPBS UPI, 1989.

Sudaryat, Yayat. Tata BasaSundaKiwari. Bandung: YramaWidya, 2007.

Thongsin, Nathaporn. A Study of English Pronunciation Problem for Thai Learners.Dissertation. Bangkok: Naresuan University, 2007.

Van Syoc, Wayland Bryce. The Phonology and Morphology of The Sundanese Language. Dissertation. Ann Arbor: University of Michigan, 1959.

APPENDIX 1

Data Matrix Of English Consonants Mispronunciation By The Native Speaker Of Sundanese

F 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 2 5 0.125 0 0 0 0 1 0 0 0 0 0 0 0 1 0 1 1 0 0 0 2 6 M [z] 0.15 0.175 I 1 1 0 0 1 1 0 0 0 1 0 0 0 1 0 0 0 1 1 2 10 0.25 F 0 1 1 1 0 2 1 2 2 2 1 2 1 1 0 2 2 2 0 2 25 0.625 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 M 21 [ʤ] 0.525 0.666666667 I 0 2 1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 2 2 1 34 0.85 F 0 2 1 2 2 2 0 2 2 2 2 2 0 2 2 2 1 2 1 2 31 0.775 0 0 1 0 1 0 0 0 0 1 0 0 0 0 0 1 0 1 0 2 7 M [ʧ] 0.175 0.583333333 I 0 2 2 2 2 0 0 2 2 1 2 2 2 2 2 2 1 2 2 2 32 0.8 F 1 2 0 0 2 0 0 0 0 1 0 0 0 2 0 0 1 1 0 2 12 0.3 0 0 0 0 1 0 0 1 0 0 0 0 0 0 1 0 0 0 0 2 5 M [ʒ] 0.125 0.141666667 I 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 F 0 2 1 2 2 0 1 1 2 1 2 2 2 2 2 1 2 2 2 2 31 0.775 1 1 1 2 1 1 0 2 2 1 2 1 2 2 2 2 2 2 1 2 M 30 [ʃ] 0.7 0.75 I 0 2 1 0 2 0 1 1 2 1 0 2 0 2 1 1 1 2 2 2 23 0.575 F 0 0 0 0 0 1 0 0 2 2 0 0 0 1 1 0 1 2 0 2 12 0.3 0 0 1 0 1 2 0 0 0 1 0 0 0 1 0 0 0 0 1 1 8 M [θ] 0.2 0.25 I 0 2 1 0 0 2 0 0 0 1 0 0 0 2 0 0 0 0 0 2 10 0.25 F 0 0 1 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 3 0.075 0 0 2 0 1 1 0 0 1 2 0 0 0 1 0 0 1 0 0 2 M 11 [ð] 0.275 0.166666667 I 0 1 0 0 0 0 0 0 0 0 1 0 0 2 0 0 0 0 0 2 6 0.15 F 2 2 1 1 0 0 0 0 0 2 1 0 0 1 1 0 1 2 0 2 16 0.4 2 1 2 2 1 1 0 2 1 2 2 1 0 2 2 2 2 1 0 2 M 28 [v] 0.7 0.466666667 I 2 2 0 1 0 0 0 1 0 0 0 2 0 0 0 1 0 0 1 2 12 0.3 F 2 2 1 2 2 2 2 2 2 2 2 2 0 2 2 2 2 2 1 2 36 0.9 1 1 1 1 1 1 2 1 1 1 0 1 1 1 1 2 2 1 1 2 M 23 [f] 0.7 0.575 I 1 1 1 1 1 1 1 1 2 1 1 2 1 2 1 1 2 1 1 2 25 0.625 % NO TOTAL TOTAL % TOTAL 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20